profile
viewpoint
Ankush Agarwal ankushagarwal San Francisco

ankushagarwal/awesome-docker 2

A curated list of Docker resources

ankushagarwal/Autopilot-TensorFlow 1

A TensorFlow implementation of this Nvidia paper: https://arxiv.org/pdf/1604.07316.pdf with some changes

ankushagarwal/awesome-ruby 1

A collection of awesome Ruby libraries, tools, frameworks and software

ankushagarwal/coursera-ml-solutions 1

Solutions for the Stanford Coursera CS229 Machine Learning Programming Assignments

ankushagarwal/ankushagarwal.github.io 0

ankushagarwal.github.io

ankushagarwal/baby-name-scraper 0

Crawls a popular Indian baby names website and dumps all names into a single file

ankushagarwal/batch-predict 0

Repository for batch predict

ankushagarwal/bootstrap 0

The most popular front-end framework for developing responsive, mobile first projects on the web.

ankushagarwal/cadvisor 0

Analyzes resource usage and performance characteristics of running containers.

ankushagarwal/caffe2-operator 0

Experimental repository for a caffe2 operator

startedijjk/notion-blog

started time in a day

startedandrewp-as-is/chrome-bookmarks.py

started time in 2 days

startedpretzelhammer/rust-blog

started time in 2 days

pull request commentlibra/libra

[cluster-test] Add basic json rpc health check when waiting for cluster health

@bors-libra r=andll

andll

comment created time in 4 days

pull request commentlibra/libra

[cluster-test] Fix account minting into proper async fn

@bors-libra r+

andll

comment created time in 5 days

pull request commentlibra/libra

[cluster-test] Use .await when waiting for tx emitter jobs to complete

@bors-libra r+

andll

comment created time in 6 days

delete branch ankushagarwal/libra

delete branch : kube-proxy2

delete time in 7 days

delete branch ankushagarwal/libra

delete branch : kube-proxy

delete time in 7 days

pull request commentlibra/libra

[cluster-test][k8s] Use kubectl proxy to forward connections

@bors-libra r=andll

ankushagarwal

comment created time in 7 days

Pull request review commentlibra/libra

[cluster-test][k8s] Use kubectl proxy to forward connections

 pub struct ClusterSwarmKube {  impl ClusterSwarmKube {     pub async fn new() -> Result<Self> {-        let result = Config::infer().await;-        let config = result.map_err(|e| format_err!("Failed to load config: {:?}", e))?;+        // This uses kubectl proxy locally to forward connections to kubernetes api server+        Command::new("/usr/local/bin/kubectl")+            .arg("proxy")+            .spawn()+            .expect("Failed to launch kubectl proxy");

Done

ankushagarwal

comment created time in 7 days

pull request commentlibra/libra

[cluster-test][k8s] Use kubectl proxy to forward connections

Running the same version fixed the issue

ankushagarwal

comment created time in 7 days

push eventankushagarwal/libra

Ankush Agarwal

commit sha 2a0fb38461c646fab5aed5a6b73ad29183747f4e

[cluster-test][k8s] Use kubectl proxy to forward connections

view details

push time in 7 days

push eventankushagarwal/libra

dependabot-preview[bot]

commit sha 6978ce25614e798c628688520e9d3ae9800c104a

Bump proc-macro2 from 1.0.12 to 1.0.13 Bumps [proc-macro2](https://github.com/alexcrichton/proc-macro2) from 1.0.12 to 1.0.13. - [Release notes](https://github.com/alexcrichton/proc-macro2/releases) - [Commits](https://github.com/alexcrichton/proc-macro2/compare/1.0.12...1.0.13) Signed-off-by: dependabot-preview[bot] <support@dependabot.com> Closes: #3932 Approved by: rexhoffman

view details

rustielin

commit sha f57f80b74f85ba653d17aefdaab4e1c6843ea272

[tf] switch on separate storage device Closes: #3937 Approved by: sausagee

view details

Sam Blackshear

commit sha ec5f0c633deb97314a4046877b877f174c4f1759

[move core modules] include sender address and amount in travel rule message Travel rule transactions will now check a signature on the message ref_id | sender_address | amount | domain_separator againt the compliance_public_key published under the recipient VASP account. The expected encodings of each of these values is: ref_id: LCS vector<u8> sender_address: LCS account address amount: LCS u64 domain_separator: UTF8 string @$LIBRA_ATTEST$@ This PR: - extends the signature checking logic to handle this larger message - changes the name of travel_rule_public_key to compliance_key, since this key will be used to sign both KYC and travel rule info - adds Move logic for rotating this key - adds transaction scripts and transaction-builder wrappers for rotation Closes: #3766 Approved by: tzakian

view details

Runtian Zhou

commit sha 787fe4c29c02d1dfa024dd29acd7c624f0301418

[move-core] Remove bytes dependency Closes: #3938 Approved by: davidiw

view details

David Dill

commit sha 552f2f7079171f4d68755e0c0e21450a2cfde8f3

[move-prover] Quantification of addresses in specs Prior to this extension, we had quantification over finite ranges and vectors. So, you could say all(1..10, |i| i < 20) (universal quantification) or any(1..10, |i| < 3) (existential quantification). This extension allows formulas to be quantified over all addresses. For example, the property: all(addresses(), |a| exists<R1>(a) ==> !exists<R2>(a)) says that no address with an R1 resource may have an R2 resource, and exists(addresses(), exists<R1>(a)) says that there is some address with an R1 resource. Such properties are often useful in module invariants to assert that a top-level resource exists at at most one address, that top-level resources are mutually exclusive, that if there is resource R1, there is always a resource R2 at the same address, etc. IMPLEMENTATION: I added a new type "addresses" representing the set of all addresses, and a new native function addresses(). When the translator encounter an "any" or "all" where the first argument is the "addresses" type, e.g. all(addresses(), |a| P(a)), it generates Boogie code: forall a : Value :: is#address(a) ==> <boogie for P(a)> For any(addresses(), |a| P(a)), it generates Boogie code: exists a : Value :: is#address(a) && <boogie for P(a)> There is a new test file tests/sources/functional/address_quant.move Closes: #3926 Approved by: wrwg

view details

Ankush Agarwal

commit sha e3cf09ff994f6910341f999f15b65e2adc4d48b3

[cluster-test][k8s] Use kubectl proxy to forward connections

view details

push time in 7 days

pull request commentlibra/libra

[cluster-test][k8s] Use kubectl proxy to forward connections

Could be the reason. I'm using devnet tag for validators.

ankushagarwal

comment created time in 7 days

create barnchankushagarwal/libra

branch : kube-proxy2

created branch time in 7 days

pull request commentlibra/libra

[cluster-test][k8s] Use kubectl proxy to forward connections

For some reason this is failing cluster-test. I am not sure if it is related to this change or not

INFO 2020-05-18 21:29:04 testsuite/cluster-test/src/cluster_swarm/cluster_swarm_kube.rs:55 Running healthcheck on http://127.0.0.1:8001
Starting to serve on 127.0.0.1:8001
INFO 2020-05-18 21:29:06 testsuite/cluster-test/src/cluster_swarm/cluster_swarm_kube.rs:55 Running healthcheck on http://127.0.0.1:8001
INFO 2020-05-18 21:29:06 testsuite/cluster-test/src/cluster_swarm/cluster_swarm_kube.rs:57 Healthcheck passed
INFO 2020-05-18 21:29:06 testsuite/cluster-test/src/main.rs:392 Deploying with devnet tag for validators and fullnodes
INFO 2020-05-18 21:29:06 testsuite/cluster-test/src/aws.rs:23 Scaling to min_desired_capacity : 60, buffer: 3, asg_name: kush-k8s-testnet-validators
INFO 2020-05-18 21:29:08 testsuite/cluster-test/src/aws.rs:64 Waiting for scale-up to complete. Current size: 0, Min Desired Size: 60
INFO 2020-05-18 21:29:19 testsuite/cluster-test/src/aws.rs:64 Waiting for scale-up to complete. Current size: 0, Min Desired Size: 60
INFO 2020-05-18 21:29:30 testsuite/cluster-test/src/aws.rs:64 Waiting for scale-up to complete. Current size: 0, Min Desired Size: 60
INFO 2020-05-18 21:29:41 testsuite/cluster-test/src/aws.rs:64 Waiting for scale-up to complete. Current size: 0, Min Desired Size: 60
INFO 2020-05-18 21:29:52 testsuite/cluster-test/src/aws.rs:64 Waiting for scale-up to complete. Current size: 0, Min Desired Size: 60
INFO 2020-05-18 21:30:03 testsuite/cluster-test/src/aws.rs:64 Waiting for scale-up to complete. Current size: 0, Min Desired Size: 60
INFO 2020-05-18 21:30:13 testsuite/cluster-test/src/aws.rs:64 Waiting for scale-up to complete. Current size: 63, Min Desired Size: 60
INFO 2020-05-18 21:30:13 testsuite/cluster-test/src/aws.rs:76 Scale up completed
INFO 2020-05-18 21:30:23 testsuite/cluster-test/src/main.rs:422 Deployment complete
INFO 2020-05-18 21:30:23 testsuite/cluster-test/src/main.rs:440 Discovered 30 validators and 30 fns
INFO 2020-05-18 21:30:24 testsuite/cluster-test/src/main.rs:513 Log tail thread started in 1029 ms
INFO 2020-05-18 21:30:45 testsuite/cluster-test/src/main.rs:793 Waiting for all validators to be healthy
WARN 2020-05-18 21:30:55 testsuite/cluster-test/src/health/log_tail.rs:46 1589837455085 Last event delay: 2522, pending 0
INFO 2020-05-18 21:31:20 testsuite/cluster-test/src/main.rs:814 All validators are now healthy
INFO 2020-05-18 21:31:20 testsuite/cluster-test/src/main.rs:688 Starting experiment all up
INFO 2020-05-18 21:31:20 testsuite/cluster-test/src/tx_emitter.rs:181 Will use 10 workers per AC with total 300 AC clients
INFO 2020-05-18 21:31:20 testsuite/cluster-test/src/tx_emitter.rs:186 Will create 15 accounts_per_client with total 4500 accounts
INFO 2020-05-18 21:33:51 testsuite/cluster-test/src/aws.rs:23 Scaling to min_desired_capacity : 0, buffer: 0, asg_name: kush-k8s-testnet-validators
thread 'main' panicked at 'called `Result::unwrap()` on an `Err` value: Failed to run experiment: [http://192.168.254.45:8080/] Failed to submit request: JsonRpcError { code: -32000, message: "Server error: invalid value: integer `66`, expected variant index 0 <= i < 2", data: None }', testsuite/cluster-test/src/main.rs:173:9
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
ankushagarwal

comment created time in 7 days

push eventankushagarwal/libra

Ankush Agarwal

commit sha 970e68258d41f05c6cefcb063d23712072079c2e

[cluster-test][k8s] Use kubectl proxy to forward connections

view details

push time in 7 days

Pull request review commentlibra/libra

[cluster-test][k8s] Use kubectl proxy to forward connections

 pub struct ClusterSwarmKube {  impl ClusterSwarmKube {     pub async fn new() -> Result<Self> {-        let result = Config::infer().await;-        let config = result.map_err(|e| format_err!("Failed to load config: {:?}", e))?;+        // This uses kubectl proxy locally to forward connections to kubernetes api server+        thread::spawn(|| {+            info!("Starting kubectl proxy");+            Command::new("/usr/local/bin/kubectl")

I shouldn't need it. I'll fix it.

ankushagarwal

comment created time in 7 days

push eventankushagarwal/libra

Ankush Agarwal

commit sha 0590a7ad8e888d9c72e0ffef529ec1ec4014399b

[cluster-test][k8s] Use kubectl proxy to forward connections

view details

push time in 7 days

push eventankushagarwal/libra

Ankush Agarwal

commit sha 42eefcd868b1d7600f9d4929ca59cb0330d41642

[cluster-test][k8s] Use kubectl proxy to forward connections

view details

push time in 7 days

push eventankushagarwal/libra

dependabot-preview[bot]

commit sha 69343a3d51b0afb2d319cfef1724f5c4e4e57052

Bump cc from 1.0.52 to 1.0.53 Bumps [cc](https://github.com/alexcrichton/cc-rs) from 1.0.52 to 1.0.53. - [Release notes](https://github.com/alexcrichton/cc-rs/releases) - [Commits](https://github.com/alexcrichton/cc-rs/compare/1.0.52...1.0.53) Signed-off-by: dependabot-preview[bot] <support@dependabot.com> Closes: #3900 Approved by: rexhoffman

view details

Wolfgang Grieskamp

commit sha 32264f4d8a25d19acbb3d3d89d1d710fd91883ca

[move-prover] Fixing a bug in handling of generic field borrow. The field type wasn't correctly instantiated with the type arguments to the struct from which is borrowed. Closes: #3903 Approved by: shazqadeer

view details

dependabot-preview[bot]

commit sha bf5c2c9b81444d6b3079c0cb7c175028beab008d

Bump anyhow from 1.0.30 to 1.0.31 Bumps [anyhow](https://github.com/dtolnay/anyhow) from 1.0.30 to 1.0.31. - [Release notes](https://github.com/dtolnay/anyhow/releases) - [Commits](https://github.com/dtolnay/anyhow/compare/1.0.30...1.0.31) Signed-off-by: dependabot-preview[bot] <support@dependabot.com> Closes: #3909 Approved by: rexhoffman

view details

Rex Hoffman

commit sha 90e79f88141a3cc7fb4b16c89f6f7f8f4d5bd0df

[testing] fix safety-rules testing Closes: #3908 Approved by: metajack

view details

Tim Zakian

commit sha 60c742a291cea9e8762609d3fb4cb0912583d74f

Add per-txn gas specifier Closes: #3902 Approved by: thefallentree

view details

Tim Zakian

commit sha 71f14dd6e878ffb0a61d73ddc779cc813fdc1ca6

Remove currency specified from LibraAccount::T Closes: #3902 Approved by: thefallentree

view details

David Wong

commit sha 1a18b6583c2ff427a224f4f114d1a58119ce9972

[fuzzing] adding storage structured fuzzer This PR adds a new structured fuzzer for storage. Meaning we can now fuzz functions that expect a struct as input (as opposed to a bytearray). The way it works is that the fuzzer gives the function a random string of bytes, and the function pretends that the random bytearray is the ouptut of an RNG it can use to construct the struct. This is called a "passthrough" RNG and we use the proptest library to build such structs from random data. The current behavior of a passthrough RNG (in the proptest library we depend on) is to crash if the input is not large enough to build the struct. This is not ideal, so to make it work the PR also patches proptest to make pass-through RNGs return 0s after they depleted the input given to them. Closes: #3872 Approved by: msmouse

view details

Sherry Xiao

commit sha f417836adc2d9b7c9cb80ee06300e17dfd998c55

[logging] Bind logging UDP socket to local port Closes: #3911 Approved by: mgorven

view details

Rex Hoffman

commit sha f652c1712c429086cbb87abf45b4c59aa94457c7

[testing] correct features to allow tests to pass Closes: #3906 Approved by: metajack

view details

Rex Hoffman

commit sha 060b5b51f51ec8261a9cda3d6a991de8acffbcfc

fixup Closes: #3906 Approved by: metajack

view details

Rex Hoffman

commit sha 6fb484606b1e19b27ae0e192c37b73bcbd412950

[testcompile] fix cargo toml to allow test compilation Closes: #3907 Approved by: metajack

view details

Mathieu Baudet

commit sha 9cbcf47d6474ebeccd0f7a3cf24e9d8ad11065f0

[network] Fix seed collision in Discovery protocol Closes: #3895 Approved by: huitseeker

view details

Mathieu Baudet

commit sha 2540bd2a3420d9890b3199a6ff133a05e1bf7e7b

[crypto] improve inline docs Closes: #3895 Approved by: huitseeker

view details

Mathieu Baudet

commit sha 3418cc5eaca5edb027e41a1e6842f87a5b9871a3

[crypto] hide DefaultHasher Closes: #3895 Approved by: huitseeker

view details

Todd Nowacki

commit sha 34f468a4d8c2d4df7d281ce4eea0fe48bf1d77aa

[move vm] Add move_to that uses signer - Added `MoveTo` and `MoveToGeneric` that operate over a `signer` instead of the `GetTxnSenderAddress` - This (plus a new native function) will open up refactors in LibraAccount Closes: #3897 Approved by: sblackshear

view details

Junkil Park

commit sha ee4a1cf829e614b8a9dac37e69e20714559c6dad

A stdlib snapshot for CAV2020 - Updated `tests/sources/stdlib/modules` for a snapshot for CAV2020 Closes: #3912 Approved by: sblackshear

view details

Clark Barrett

commit sha 6798b1cd50ac7d524d3e494783910b3d7e827eef

Replaced type alias "Edge" - replaced with "int" Closes: #3915 Approved by: cbarrettfb

view details

Todd Nowacki

commit sha f35e0f13345f1f862a9082f4393891ad4e5ab0e2

[move source language] Add support for module member aliases - Refactored parser::ast Module Members - Refactored aliases and unscoped names - Modules members are now introduced as an implicit alias, much like Self - Removed unscoped name logic from naming - You can now use module members from other modules, much like Rust Closes: #3920 Approved by: bob-wilson

view details

Tim Zakian

commit sha 2d691b5a662bd541b18214df2353d56518a3b9e8

Add transaction fee distribution per-currency Closes: #3916 Approved by: runtian-zhou

view details

aldenhu

commit sha 415d20322ec0da9bfda4b35e716a22622d891300

[backup] add state_snapshot_chunk endpoint stub and test routing Closes: #3890 Approved by: lightmark

view details

push time in 7 days

push eventankushagarwal/one-click-hugo-cms

Aditya Jain

commit sha 4efa5cd2c3e2133fe9136288511f6d624c5460bf

Update Pages “home”

view details

push time in 8 days

push eventankushagarwal/one-click-hugo-cms

Ankush Agarwal

commit sha e3dd560018af260827ae39a00a22a26dbbecfd98

Update

view details

push time in 8 days

push eventankushagarwal/one-click-hugo-cms

Ankush Agarwal

commit sha d301de40e73b66d4810539d38523b95abebb4b12

Update

view details

push time in 8 days

push eventankushagarwal/one-click-hugo-cms

Ankush Agarwal

commit sha ab3fd860fd44467d89bad5ac51007e98b7311215

Update

view details

push time in 8 days

push eventankushagarwal/one-click-hugo-cms

Ankush Agarwal

commit sha da9247eb4841299ef434cad0afecc9ceac9940ee

Update

view details

push time in 8 days

push eventankushagarwal/one-click-hugo-cms

Ankush Agarwal

commit sha 690bb4f7f2c89e78b9b4364062b925c340a4881e

Update

view details

push time in 8 days

push eventankushagarwal/one-click-hugo-cms

Ankush Agarwal

commit sha d134a830107c17cc0ebbb54a5307c6eff7d819bc

Update

view details

push time in 8 days

create barnchankushagarwal/one-click-hugo-cms

branch : master

created branch time in 8 days

created repositoryankushagarwal/one-click-hugo-cms

created time in 8 days

push eventankushagarwal/ankushagarwal.github.io

Ankush Agarwal

commit sha cc2d7b4932bac36019461387233dae1a7f636bf6

Delete old site - redirect to LinkedIn

view details

push time in 8 days

push eventankushagarwal/libra

Ankush Agarwal

commit sha 29aab75846d91b9cafcc7b93bd86302521f4982b

[cluster-test][k8s] Use kubectl proxy to forward connections

view details

push time in 11 days

push eventankushagarwal/libra

Junkil Park

commit sha d1cb337499e74a2dbe2efbfb459793a0d90dc571

Verified standard Vector module - specified `reverse` function and verified against the boogie model - verified new functions in the updated vector module such as `singleton` and `index_of` - added the builtin boogie model for `Vector::index_of` - added test cases in `verify_vector.move` and `simple_vector_client.move` - fixed the remaining bug that reverses the order of multiple return values of the *generic call* - Turned off the performance_200511 regression test temporally (i.e., pragma verify=false ) - There are two ISSUES: - (1) there is a weired behavior around the builtin boogie functiob `contains_vector`. See FIXMEs in `simple_vector_client.move`. - (2) Z3 performance problem rises again. This new builtin boogie procedure `$Vector_index_of` sometimes (not always) make the test (performance_200511) very slow (> 10 mins) or hang although this is not used in the test script (performance_200511). This issue is on and off. See FIXMEs in Prelude. Closes: #3883 Approved by: wrwg

view details

Zihao Chen

commit sha 3943294a8c060edf61f58614a4ec3017bd72d07f

[cluster_test] Client side latency statistics Closes: #3891 Approved by: andll

view details

Ankush Agarwal

commit sha fca22b8248d73a9c59453a8e7dd045e32f57a755

[cluster-test][k8s] Use kubectl proxy to forward connections

view details

push time in 11 days

push eventankushagarwal/libra

Ankush Agarwal

commit sha 309a54ccb444cbb1bdf3f780a5bb906ebf6e9fb8

[cluster-test][k8s] Use kubectl proxy to forward connections

view details

push time in 11 days

push eventankushagarwal/libra

Ankush Agarwal

commit sha 1e67f8c442de5d1828bd530d5f5693ee55f05e15

[cluster-test][k8s] Use kubectl proxy to forward connections

view details

push time in 11 days

push eventankushagarwal/libra

Ankush Agarwal

commit sha f387d4e97a5bf4e62aaf94bfcdcb1fce51dcc491

[cluster-test][k8s] Use kubectl proxy to forward connections

view details

push time in 11 days

push eventankushagarwal/libra

Ankush Agarwal

commit sha d72e8170cc38cb62fe2a5eca5e5921bc3fba4cfb

[cluster-test][k8s] Use kubectl proxy to forward connections

view details

push time in 11 days

push eventankushagarwal/libra

Ankush Agarwal

commit sha d12a6c1148eb5cac007a35fbee0db532d6d3c319

[cluster-test][k8s] Use kubectl proxy to forward connections

view details

push time in 11 days

PR opened libra/libra

Reviewers
[cluster-test][k8s] Use kubectl proxy to forward connections cluster_test

Summary

Use kubectl proxy to forward connections

Test Plan

WIP

+11 -3

0 comment

2 changed files

pr created time in 11 days

create barnchankushagarwal/libra

branch : kube-proxy

created branch time in 11 days

push eventankushagarwal/libra

David Wong

commit sha f008c1abaeb717e00a7616f58827ac6f9442118a

[fuzzing] coverage needs to run with fuzzing flag enabled Closes: #3842 Approved by: huitseeker

view details

Victor Gao

commit sha 50daf57d9d75a9fc804bf2b9a9c4116af3943df3

[language] move transaction argument from libra_types into move_core_types Closes: #3869 Approved by: dariorussi

view details

rustielin

commit sha 4f2f9dc8bf9a0b92a17a71fc3e4e60ce6ba86cec

[tf] fix log paths for nonpersistent logrotation Closes: #3864 Approved by: sausagee

view details

David Wolinsky

commit sha 8b57364eb625d8bbf7aadc768c684eff4f32841c

[swarm] some cleanups Was reading the code to prepare for potential management smoke tests Closes: #3866 Approved by: joshlind

view details

Runtian Zhou

commit sha 2a9b414aef5a39af5794e5012477184486e6b74f

[libra_vm] cleanup dependency for LibraVM Closes: #3867 Approved by: sblackshear

view details

David Wolinsky

commit sha d7befa3fa5e1ed4b182a236c8ce855ab5acafd09

[config] Merge service and libranet port randomization Not really sure why these are separated since we likely need them randomized for the same reasons. Closes: #3870 Approved by: phlip9

view details

Todd Nowacki

commit sha 1695682be7840a20639b7bd56c7cf97fc63afa78

[move vm] Added signer type - Added the primitive `signer` type as described in #3679 - Name is `signer` instead of `sender` based on in person discussions - Operations over the type will come in a later PR - Long term this system will replace `MoveToSender` and `GetTxnSenderAddress` Closes: #3865 Approved by: sblackshear

view details

Victor Gao

commit sha 5057a7246c26084849611a88c3ba236bb5e507df

[language] move type tag parser from functional_tests into move_core_types Closes: #3871 Approved by: dariorussi

view details

rustielin

commit sha 2f1847351d638c1b5245a3c23ca331f0daef47a6

[tf] hotfix logrotate for nonpersistent Closes: #3873 Approved by: sausagee

view details

dependabot-preview[bot]

commit sha 4afecabf6c4476536529e8f7ae4e0d76fa28106d

Bump backtrace from 0.3.47 to 0.3.48 Bumps [backtrace](https://github.com/rust-lang/backtrace-rs) from 0.3.47 to 0.3.48. - [Release notes](https://github.com/rust-lang/backtrace-rs/releases) - [Commits](https://github.com/rust-lang/backtrace-rs/compare/0.3.47...0.3.48) Signed-off-by: dependabot-preview[bot] <support@dependabot.com> Closes: #3875 Approved by: rexhoffman

view details

dependabot-preview[bot]

commit sha 7729e2e1029dbc65cd1cfb7885e547252de7f408

Bump anyhow from 1.0.29 to 1.0.30 Bumps [anyhow](https://github.com/dtolnay/anyhow) from 1.0.29 to 1.0.30. - [Release notes](https://github.com/dtolnay/anyhow/releases) - [Commits](https://github.com/dtolnay/anyhow/compare/1.0.29...1.0.30) Signed-off-by: dependabot-preview[bot] <support@dependabot.com> Closes: #3876 Approved by: rexhoffman

view details

dependabot-preview[bot]

commit sha 51634278018a426cd4c9ecf0301ea85f95efe72d

Bump once_cell from 1.3.1 to 1.4.0 Bumps [once_cell](https://github.com/matklad/once_cell) from 1.3.1 to 1.4.0. - [Release notes](https://github.com/matklad/once_cell/releases) - [Changelog](https://github.com/matklad/once_cell/blob/master/CHANGELOG.md) - [Commits](https://github.com/matklad/once_cell/compare/v1.3.1...v1.4.0) Signed-off-by: dependabot-preview[bot] <support@dependabot.com> Closes: #3877 Approved by: rexhoffman

view details

David Wolinsky

commit sha eb904f942c5663ce8508cb1be4fc4ab71759dfb0

[management] fixup constants Prefer not to be importing constants all over the place. Also due to some feature hacks, couldn't run the test locally, now we can. Finally, added an intermediary for layouts to strings. Closes: #3878 Approved by: joshlind

view details

David Wolinsky

commit sha 683af00337d5d2691ad0762412318b5f5381a433

[secure-storage] Swap toml for json Toml is not a very good serialization for complex data storage that we're doing in secure-store. In fact, I found it cannot serialize a transaction. So we're moving over to json that should be slightly more readable and probably more compact. Closes: #3878 Approved by: joshlind

view details

David Wolinsky

commit sha 286088c0f2048c59a4475417efb12b3f8e23a8b6

[management] rewrite tests to use ondisk storage This is an incremental step to allowing this framework to be used for integration tests. When the original library was written, ondisk storage couldn't be used due to lack of namespacing. Now it has namespacing and can be used. This allows us to remove the ignore tag and also use these frameworks for other types of tests, which is the motivation for StorageHelper. Closes: #3878 Approved by: joshlind

view details

aldenhu

commit sha a55c852079ff0a22c6eba23b5fbd3a458e947115

[storage] trivial: create storage/backup folder and move exsiting cli over Closes: #3803 Approved by: wqfish

view details

aldenhu

commit sha 27ecfb1265d700a18b278077181fcc24b063a360

HTTP BackupService Closes: #3803 Approved by: wqfish

view details

aldenhu

commit sha 1c1db58a1a15993e9207b369c10c23e7fe26e28d

[storage] backup-cli integration test Closes: #3803 Approved by: wqfish

view details

aldenhu

commit sha 97c458e54545e01a7f04ee28ea27164368ea20ce

[storage] use HTTP backup service (for one API) Closes: #3803 Approved by: wqfish

view details

Zekun Li

commit sha 774a7c245b609fa8eae0776c6b3664634da2ebe2

[consensus] twins bootstrap Bring back SMRNode to unblock twins integration. Closes: #3886 Approved by: zekun000

view details

push time in 11 days

PR closed libra/libra

Reviewers
Bump k8s-openapi from 0.7.1 to 0.8.0 cla-signed dependencies

Bumps k8s-openapi from 0.7.1 to 0.8.0. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/Arnavion/k8s-openapi/releases">k8s-openapi's releases</a>.</em></p> <blockquote> <h2>v0.8.0</h2> <ul> <li> <p>BREAKING CHANGE: Support for v1.8, v1.9 and v1.10 API servers has been dropped. These versions became hard to test with <code>kubectl</code> and <code>kind</code> are are not supported by major cloud providers.</p> </li> <li> <p>BREAKING CHANGE: <code>k8s_openapi::apiextensions_apiserver::pkg::apis::apiextensions::v1::JSONSchemaPropsOrArray</code>, <code>JSONSchemaPropsOrBool</code> and <code>JSONSchemaPropsOrStringArray</code> types now wrap the <code>v1::JSONSchemaProps</code> type. Previously they incorrectly wrapped the <code>v1beta1::JSONSchemaProps</code> type.</p> </li> <li> <p>BREAKING CHANGE: Turning the <code>api</code> feature off now also disables the <code>k8s_openapi::{http,percent_encoding,url}</code> re-exports, the <code>k8s_openapi::percent_encoding2</code> module, the <code>k8s_openapi::{RequestError,ResponseError,ResponseBody}</code> types, the <code>k8s_openapi::Response</code> trait, and the <code>k8s_openapi::{Create,Delete,List,Patch,Replace,Watch}{Optional,Response}</code> types. All of these were only used by the API functions which had been disabled by turning the <code>api</code> feature off.</p> </li> <li> <p>BUGFIX: <code>k8s_openapi::apimachinery::pkg::apis::meta::v1::MicroTime</code> and <code>Time</code> are now serialized with six and zero decimals in their seconds respectively, to match the API server's expectations. Previously they would be serialized with a variable number of decimals up to nine.</p> </li> <li> <p>FEATURE: Added support for Kubernetes 1.18 under the <code>v1_18</code> feature.</p> </li> <li> <p>FEATURE: <code>k8s_openapi::ByteString</code> now impls <code>PartialOrd</code> and <code>Ord</code>.</p> </li> <li> <p>FEATURE: <code>k8s_openapi::apimachinery::pkg::apis::meta::v1::MicroTime</code> and <code>Time</code> now impl <code>Eq</code>, <code>PartialOrd</code> and <code>Ord</code>.</p> </li> <li> <p>FEATURE: If your crate has a dependency on <code>k8s-openapi</code>, you can now use a build script to detect which version feature has been enabled on the crate. This is a more verbose but also more flexible alternative to using the <code>k8s_if_</code> version detection macros in your crate code.</p> </li> <li> <p>FEATURE: <code>k8s-openapi-derive</code>'s <code>#[derive(CustomResourceDefinition)]</code> now supports emitting a subresources field in the generated CR type.</p> </li> </ul> <p>Corresponding Kubernetes API server versions:</p> <ul> <li>v1.11.10</li> <li>v1.12.10</li> <li>v1.13.12</li> <li>v1.14.10</li> <li>v1.15.11</li> <li>v1.16.9</li> <li>v1.17.5</li> <li>v1.18.2</li> </ul> </blockquote> </details> <details> <summary>Changelog</summary> <p><em>Sourced from <a href="https://github.com/Arnavion/k8s-openapi/blob/master/CHANGELOG.md">k8s-openapi's changelog</a>.</em></p> <blockquote> <h1>v0.8.0 (2020-05-02)</h1> <ul> <li> <p>BREAKING CHANGE: Support for v1.8, v1.9 and v1.10 API servers has been dropped. These versions became hard to test with <code>kubectl</code> and <code>kind</code> are are not supported by major cloud providers.</p> </li> <li> <p>BREAKING CHANGE: <code>k8s_openapi::apiextensions_apiserver::pkg::apis::apiextensions::v1::JSONSchemaPropsOrArray</code>, <code>JSONSchemaPropsOrBool</code> and <code>JSONSchemaPropsOrStringArray</code> types now wrap the <code>v1::JSONSchemaProps</code> type. Previously they incorrectly wrapped the <code>v1beta1::JSONSchemaProps</code> type.</p> </li> <li> <p>BREAKING CHANGE: Turning the <code>api</code> feature off now also disables the <code>k8s_openapi::{http,percent_encoding,url}</code> re-exports, the <code>k8s_openapi::percent_encoding2</code> module, the <code>k8s_openapi::{RequestError,ResponseError,ResponseBody}</code> types, the <code>k8s_openapi::Response</code> trait, and the <code>k8s_openapi::{Create,Delete,List,Patch,Replace,Watch}{Optional,Response}</code> types. All of these were only used by the API functions which had been disabled by turning the <code>api</code> feature off.</p> </li> <li> <p>BUGFIX: <code>k8s_openapi::apimachinery::pkg::apis::meta::v1::MicroTime</code> and <code>Time</code> are now serialized with six and zero decimals in their seconds respectively, to match the API server's expectations. Previously they would be serialized with a variable number of decimals up to nine.</p> </li> <li> <p>FEATURE: Added support for Kubernetes 1.18 under the <code>v1_18</code> feature.</p> </li> <li> <p>FEATURE: <code>k8s_openapi::ByteString</code> now impls <code>PartialOrd</code> and <code>Ord</code>.</p> </li> <li> <p>FEATURE: <code>k8s_openapi::apimachinery::pkg::apis::meta::v1::MicroTime</code> and <code>Time</code> now impl <code>Eq</code>, <code>PartialOrd</code> and <code>Ord</code>.</p> </li> <li> <p>FEATURE: If your crate has a dependency on <code>k8s-openapi</code>, you can now use a build script to detect which version feature has been enabled on the crate. This is a more verbose but also more flexible alternative to using the <code>k8s_if_</code> version detection macros in your crate code.</p> </li> <li> <p>FEATURE: <code>k8s-openapi-derive</code>'s <code>#[derive(CustomResourceDefinition)]</code> now supports emitting a subresources field in the generated CR type.</p> </li> </ul> <p>Corresponding Kubernetes API server versions:</p> <ul> <li>v1.11.10</li> <li>v1.12.10</li> <li>v1.13.12</li> <li>v1.14.10</li> <li>v1.15.11</li> <li>v1.16.9</li> <li>v1.17.5</li> <li>v1.18.2</li> </ul> <hr /> </blockquote> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/Arnavion/k8s-openapi/commit/48023effff2e9ef2fea43d520a2c65d29f2164e2"><code>48023ef</code></a> v0.8.0</li> <li><a href="https://github.com/Arnavion/k8s-openapi/commit/3d1e64a1a6dca795945fc526f69a825fbcce1407"><code>3d1e64a</code></a> <code>test.sh create-cluster</code> needs the image directory.</li> <li><a href="https://github.com/Arnavion/k8s-openapi/commit/4f89bbac8050b2cb8d67d1414f9bd530451e7462"><code>4f89bba</code></a> Fix some unnecessary uses of std::mem::replace</li> <li><a href="https://github.com/Arnavion/k8s-openapi/commit/797b3ae7a4418610f06f9d42b644f6311317920a"><code>797b3ae</code></a> Fix and suppress new clippy lints.</li> <li><a href="https://github.com/Arnavion/k8s-openapi/commit/8fa86655ffd317ed0d631f49f4ea7f8c77eefabe"><code>8fa8665</code></a> Add note about how to clean up after <code>kind build-node-image</code></li> <li><a href="https://github.com/Arnavion/k8s-openapi/commit/93da22fc701723228b2444da828ff9200e31e63f"><code>93da22f</code></a> Updated v1.16 to v1.16.9, v1,17 to v1,17.5, v1.18 to v1.18.2</li> <li><a href="https://github.com/Arnavion/k8s-openapi/commit/a26c3eac7b1e91f3894001a3afe5ef51cfb747b0"><code>a26c3ea</code></a> Update v1.18 to v1.18.1</li> <li><a href="https://github.com/Arnavion/k8s-openapi/commit/e6cbbf07eade671a5f3eb1b449967765ce1a9a74"><code>e6cbbf0</code></a> Stop using self_link since it's deprecated.</li> <li><a href="https://github.com/Arnavion/k8s-openapi/commit/39b7b2d89c0e048da92f37e0827919a305180f75"><code>39b7b2d</code></a> Identify the specific replay that fails the test.</li> <li><a href="https://github.com/Arnavion/k8s-openapi/commit/2a79c8ec42c945b691b11147979afd570e314f84"><code>2a79c8e</code></a> Clean images gunk by exporting them after they're built.</li> <li>Additional commits viewable in <a href="https://github.com/Arnavion/k8s-openapi/compare/v0.7.1...v0.8.0">compare view</a></li> </ul> </details> <br />

Dependabot compatibility score

You can trigger a rebase of this PR by commenting @dependabot rebase.


<details> <summary>Dependabot commands and options</summary> <br />

You can trigger Dependabot actions by commenting on this PR:

  • @dependabot rebase will rebase this PR
  • @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
  • @dependabot merge will merge this PR after your CI passes on it
  • @dependabot squash and merge will squash and merge this PR after your CI passes on it
  • @dependabot cancel merge will cancel a previously requested merge and block automerging
  • @dependabot reopen will reopen this PR if it is closed
  • @dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
  • @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
  • @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
  • @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)
  • @dependabot use these labels will set the current labels as the default for future PRs for this repo and language
  • @dependabot use these reviewers will set the current reviewers as the default for future PRs for this repo and language
  • @dependabot use these assignees will set the current assignees as the default for future PRs for this repo and language
  • @dependabot use this milestone will set the current milestone as the default for future PRs for this repo and language
  • @dependabot badge me will comment on this PR with code to add a "Dependabot enabled" badge to your readme

Additionally, you can set the following in your Dependabot dashboard:

  • Update frequency (including time of day and day of week)
  • Pull request limits (per update run and/or open at any time)
  • Out-of-range updates (receive only lockfile updates, if desired)
  • Security updates (receive only security updates, if desired)

</details>

+16 -2

4 comments

2 changed files

dependabot-preview[bot]

pr closed time in 12 days

pull request commentlibra/libra

Bump k8s-openapi from 0.7.1 to 0.8.0

Thanks @rexhoffman

Based on our current usage, we don't need the RefreshableToken feature.

dependabot-preview[bot]

comment created time in 12 days

push eventankushagarwal/libra

jolestar

commit sha 6dee1c8caaa894d50fbd2518809fa036068b34cd

[serde] Fix serde_json::from_value error #3613 Closes: #3614 Approved by: davidiw

view details

dependabot-preview[bot]

commit sha 664cea72c64e54909b4881002d6bda65f58ce97c

Bump proc-macro2 from 1.0.10 to 1.0.12 Bumps [proc-macro2](https://github.com/alexcrichton/proc-macro2) from 1.0.10 to 1.0.12. - [Release notes](https://github.com/alexcrichton/proc-macro2/releases) - [Commits](https://github.com/alexcrichton/proc-macro2/compare/1.0.10...1.0.12) Signed-off-by: dependabot-preview[bot] <support@dependabot.com> Closes: #3712 Approved by: rexhoffman

view details

dependabot-preview[bot]

commit sha d90229104d19e473aa22eb33f4094dd40d6fdee6

Bump serde_bytes from 0.11.3 to 0.11.4 Bumps [serde_bytes](https://github.com/serde-rs/bytes) from 0.11.3 to 0.11.4. - [Release notes](https://github.com/serde-rs/bytes/releases) - [Commits](https://github.com/serde-rs/bytes/compare/0.11.3...0.11.4) Signed-off-by: dependabot-preview[bot] <support@dependabot.com> Closes: #3710 Approved by: wqfish

view details

Zekun Li

commit sha 290138d59826d0eb57aa87a65c801a9d9ec79ecd

[consensus] cleanup last_vote_sent Closes: #3704 Approved by: davidiw

view details

Christina Lee

commit sha a405b894cdbe3b6e6bed1e7da222c919d9573762

[mempool] removed broadcast ticker Closes: #3715 Approved by: phoenix-antigravity

view details

Christina Lee

commit sha a4b3bbf14698ce34f217c4f8801274d1b6f0c8c0

[mempool] unified coordinators Closes: #3715 Approved by: phoenix-antigravity

view details

Christina Lee

commit sha 6341fc3ce518dd7a8e4f9d3f42d80963335fa6a0

[mempool] method signature and clone cleanup Closes: #3715 Approved by: phoenix-antigravity

view details

François Garillot

commit sha 04dbc38bdeb7c59c5a7ad59ba5437f31931b1275

Removing hex from config Closes: #3660 Approved by: metajack

view details

François Garillot

commit sha 990d909b3724fabce2754fe6eba07e4c9b7d5a45

Removing once_cell from onchain-discovery Closes: #3660 Approved by: metajack

view details

François Garillot

commit sha 6f6c992821ce5d0a60ed0dd32b68cbb29029cc5c

Removing serde from storage-client Removing lcs from storage-client Closes: #3660 Approved by: metajack

view details

François Garillot

commit sha 1d7d4111a5b0bb91a3a355059217644e3f4505ab

Removing proptest from mempool Removing rand from mempool Closes: #3660 Approved by: metajack

view details

François Garillot

commit sha e2aca2eac63ea0d6accb1a1447f7ddec6038745b

Removing libradb from consensus Closes: #3660 Approved by: metajack

view details

François Garillot

commit sha b1e83e0ada8f2e02b797e8cf297984909189b999

Removing hex from consensus-types Closes: #3660 Approved by: metajack

view details

François Garillot

commit sha c50d2a4909e5d40ef845962abf0e5f7d5dec26f6

Removing futures from vm-validator Closes: #3660 Approved by: metajack

view details

François Garillot

commit sha 92d166e8ec29697343ce55c484e39ba11db58fd8

Removing chrono from cluster-test Closes: #3660 Approved by: metajack

view details

François Garillot

commit sha a8352f9c27e08a939b58b2bb01981c79c8353e73

Removing prost from admission-control-service Closes: #3660 Approved by: metajack

view details

François Garillot

commit sha c131cca3fdc0d6a2dc55d2913736e5944a532f5d

Removing anyhow from executor-benchmark Removing tokio from executor-benchmark Closes: #3660 Approved by: metajack

view details

François Garillot

commit sha c934d9c10932e48136cbc1d4b9da38878bdee834

Removing cli from libra-swarm Closes: #3660 Approved by: metajack

view details

François Garillot

commit sha f01d17040f94c19817627c6104ee6566662cd168

Removing stdlib from stackless-bytecode-generator Closes: #3660 Approved by: metajack

view details

François Garillot

commit sha 903e6c6a33836298b45d6a38459c78ed4a8414e9

Removing itertools from test-utils Closes: #3660 Approved by: metajack

view details

push time in 12 days

push eventankushagarwal/libra

Zekun Li

commit sha f28fb21ef45b6aa2ddd09c07e91bf3af5668b6a1

[epoch change] get_epoch_proof from libradb instead of state computer Since we have direct access to libradb in epoch manager now, we don't have to route through state computer. Closes: #3627 Approved by: davidiw

view details

Zekun Li

commit sha 61204087ac45e5c8a61ef4d7645c97931393b5de

[epoch change] turn VerifierType into trait This fits better with trait, simplify the impl and avoid a few clones. Closes: #3627 Approved by: davidiw

view details

David Wolinsky

commit sha 572c482ecdd7d8e3d947b79aca3a47fc007e6034

[config] Update readme Closes: #3635 Approved by: joshlind

view details

Ankush Agarwal

commit sha b5666e19883c6bc634fb6069646b576e7699989e

[cluster-test] Use correct asg name instead of hardcoded value Closes: #3636 Approved by: davidiw

view details

dependabot-preview[bot]

commit sha deee1a06bd38564b9afb2e8270081c6370ce4bcd

Bump serde_json from 1.0.51 to 1.0.52 Bumps [serde_json](https://github.com/serde-rs/json) from 1.0.51 to 1.0.52. - [Release notes](https://github.com/serde-rs/json/releases) - [Commits](https://github.com/serde-rs/json/compare/v1.0.51...v1.0.52) Signed-off-by: dependabot-preview[bot] <support@dependabot.com> Closes: #3643 Approved by: rexhoffman

view details

Joel Marcey

commit sha 2f2759b6cb4e9207bd59f6a391fdf709ca6b45e4

[RFC][Move] Change move build output directory to something less generic Closes: #3629 Approved by: tnowacki

view details

Joel Marcey

commit sha d91685f6a2ce10c1bf04eca9599462fac7df10a7

Capitalization Closes: #3629 Approved by: tnowacki

view details

Joel Marcey

commit sha 1338a917dbafecdd761d8968f933631cca5b3838

Less capitalization :) Closes: #3629 Approved by: tnowacki

view details

Dm. Yakushev

commit sha d030194c2a39e194480d02e7ecd5683ce2d1c1c2

[vm]Native function refactoring. -Make native functions pluggable. -Moved special case native functions like save_account and emit_event to others functions. -Created context for native functions, which expands their capabilities. Closes: #3472 Approved by: dariorussi

view details

Bob Wilson

commit sha 6d658caf5a109d40dd80e30de6ba7ff2d04a21a7

[move-lang] Parser cleanup for GlobalCalls and ModuleAccess Wolfgang's changes in #3438 laid the groundwork for some simplifications around the way that we use the ModuleAccess nodes. Besides simplifying the parser, I'm looking into adding support for name aliases and plan to move toward using the ModuleAccess AST nodes to represent all the different kinds of names. (I'd like to rename "ModuleAccess" but I'm leaving that for a separate PR.) - Extend the ModuleAccess AST node to also include global names. - Distinguish Calls and GlobalCalls in the expansion phase, not the parser. - Introduce a new "NameExp" non-terminal in the grammar (instead of the parse_pack_or_call_or_generic_name helper function) and refactor the code that uses it. Closes: #3634 Approved by: tnowacki

view details

Michael Gorven

commit sha bc9e020746caf1a712037fc4bd3e0898f6050775

[tf] Add tags to all resources Tag resources with TF config and workspace to help with billing. Test Plan: `terraform apply` in my workspace Closes: #3647 Approved by: rustielin

view details

Ankush Agarwal

commit sha fe282088ade2bfcde757d1b85b37be1fb7cf4892

[cluster-test] Add a teardown method for ClusterTestRunner Closes: #3646 Approved by: andll

view details

Tim Zakian

commit sha 3f4d918e46d7591d9d7ad2d4e873dfb5e239cedd

[types] Update directory structure for account_config Closes: #3565 Approved by: dariorussi

view details

Yucong Sun

commit sha b04738f65d06e77275820f01559919f7e36557d7

Remove depdenency of json-rpc on client Closes: #3648 Approved by: phoenix-antigravity

view details

Todd Nowacki

commit sha 6b728d4ba7f8be82856099ddad74938a68ff8453

Bytecode verifier] Added (unused) structural control flow checks - Added structual control flow checks that verify loops behave as if they were written as "normal" loops - The check is not yet integrated into the code unit checker Closes: #3499 Approved by: dariorussi

view details

Philip Hayes

commit sha 27edfff6ed862dc8b6ab8f28e4a759b8a31ca2a2

[network-address] add temp multiaddr compatibility methods + Allows fallibly converting `NetworkAddress` <-> `Multiaddr` and `network_address::Protocol` <-> `parity_multiaddr::Protocol`. + This is just a temporary conversion method until we finish converting all crates to using the new `NetworkAddress` struct. Closes: #3573 Approved by: metajack

view details

Philip Hayes

commit sha 8e8b50ab4b2ea72ada29639adec0013fa8bc97ea

[network-address] convert netcore and network benches Closes: #3573 Approved by: metajack

view details

Philip Hayes

commit sha 501c64264c4f0c54c01238be7785a86fbfd8afef

[network-address] convert network, state-synchronizer, mempool Closes: #3573 Approved by: metajack

view details

Philip Hayes

commit sha 1d39c19582dcde4a4149b3de44ef37112bed0079

[network-address][breaking] convert types, onchain-discovery, genesis + Notice that the onchain data structures (in types/) are represented using `RawNetworkAddress` in Rust, which corresponds with `vector<u8>` in Move contracts. + This also allows to easily ignore `RawNetworkAddress`es that don't parse correctly, as seen in `network/onchain-discovery/src/types.rs`. Closes: #3573 Approved by: metajack

view details

Philip Hayes

commit sha f9ba92cec6cea90c7003674a0ab3417dfff04800

[network-address] convert config, config-builder Closes: #3573 Approved by: metajack

view details

push time in 22 days

delete branch ankushagarwal/libra

delete branch : cluster-test-scaledown

delete time in a month

pull request commentlibra/libra

[cluster-test] Add a teardown method for ClusterTestRunner

@bors-libra r=andll

ankushagarwal

comment created time in a month

push eventankushagarwal/libra

Shaz Qadeer

commit sha a1f9b8211f113b5939bb6e3bc162246c67497d6e

allowed dataflow analysis to compute error output for a block Closes: #3604 Approved by: junkil-park

view details

Wolfgang Grieskamp

commit sha 49cfe3d2fa19bd400b360aac0f6425eb973e4b9f

[move-prover] Refactor test suite driver and copy new lib into test tree. This refactors our test directory and driver as follows: - all top-level tests have been moved into `sources/functional` - a copy of the new standard library has been placed in `sources/new_stdlib` - tests don't need to longer declare their dependencies via the `// dep` comment. Instead, dependencies are automatically detected by using a search path. - different test directories (like functional, stdlib, etc.) can use different flags and search paths. This is configured in `testsuite.rs`, and allows for example `new_stdlib` to find dependencies at a different place than `stdlib`. - for `new_stdlib`, verification is off by default and need to be turned on using `pragma verify = true` (either in a module spec block for the whole module or per function spec block) - a `.exp` file is not longer generated if the test passes without any errors. On UPBL=1, existing .exp files are instead removed if there are no errors. This prepares us to run directly against the standard library source of truth at a later point. There is a nasty issue with native functions of the same name having different signatures in old and new library. To work around this, a hack was added to generate different boogie names so the old/new lib sources can be used by the same prelude. The native functions for the new library, as well as new vector function `index_of`, are not implemented yet and `assert false`. Closes: #3609 Approved by: DavidLDill

view details

dependabot-preview[bot]

commit sha 11ad030eea71ff63d437c03ed2ad1587b2df7295

Bump criterion from 0.3.1 to 0.3.2 Bumps [criterion](https://github.com/bheisler/criterion.rs) from 0.3.1 to 0.3.2. - [Release notes](https://github.com/bheisler/criterion.rs/releases) - [Changelog](https://github.com/bheisler/criterion.rs/blob/master/CHANGELOG.md) - [Commits](https://github.com/bheisler/criterion.rs/compare/0.3.1...0.3.2) Signed-off-by: dependabot-preview[bot] <support@dependabot.com> Closes: #3612 Approved by: rexhoffman

view details

Wolfgang Grieskamp

commit sha b54b3e9d3d36afa3dc1ab4d4dafd083ce062bd85

[move-prover] Let move prover support transaction scripts. Until this PR, we could only run the prover on Move sources which define modules. This PR adds support for scripts. Scripts are mapped to a "pseudo" module with a single function for the transactions, such that most parts of the existing code stay the same. We also needed to extend the Move parser to support spec blocks in transaction sources. During testing, the PR also discovered a bug in using schemas imported from other modules, which is fixed now. This PR copies all existing transaction scripts into tests/sources/new_stdlib/transaction_scripts, where they are by default processed as tests with the `--verify=none` flag. Closes: #3605 Approved by: DavidLDill

view details

dependabot-preview[bot]

commit sha dcd8631647200e086dbac50f31e40941ded19ba2

Bump simplelog from 0.7.5 to 0.7.6 Bumps [simplelog](https://github.com/drakulix/simplelog.rs) from 0.7.5 to 0.7.6. - [Release notes](https://github.com/drakulix/simplelog.rs/releases) - [Changelog](https://github.com/Drakulix/simplelog.rs/blob/master/CHANGELOG.md) - [Commits](https://github.com/drakulix/simplelog.rs/compare/v0.7.5...v0.7.6) Signed-off-by: dependabot-preview[bot] <support@dependabot.com> Closes: #3611 Approved by: rexhoffman

view details

Christina Lee

commit sha 5f5d7b67a795e44ca6eb18fa07955b8d3f1dfa67

[config] remove upstream peers from state sync config Closes: #3590 Approved by: davidiw

view details

Bob Wilson

commit sha dde9c1b9610375d314505781f169f3098664a356

[language] update the libra-vm monitoring counters - Adopt the same conventions for Prometheus counters in libra-vm as in the rest of Libra, with a separate name for each counter and using the same naming convention as elsewhere. - Simplify the status counters: Failed transactions should be rare, so I don't think we need to count separately for each type of failure. If there's an increase in failure counts, I expect someone would look into the logs to see what is causing that. - Change to access the counters directly from libra_vm.rs. Previously the counters were private to the counters.rs file with public functions and macros to update them. This didn't seem to add much value and it was inconsistent with other Libra components. - It's not clear whether all the time histograms here are needed but the original motivation of using them to correlate with gas costs seems still valid, so I did not remove any of them. Closes: #3600 Approved by: dariorussi

view details

Tim Zakian

commit sha c9d59085184cbc6daed56f7e83ebb92aa43d2607

[language] Update source-map to use LCS. Add source-map deserialization option to Move language compiler. Closes: #3593 Approved by: sblackshear

view details

Young Yang Liauw

commit sha e0c2faa643d686ebcc144dd535b43456a81cc6c5

[docker] use wget in build We started seeing frequent build failures, on CI hosts and local dev machines, due to a mysterious connection reset (openssl errno 104) in downloading resource from certain 3rd party sites curl: (56) OpenSSL SSL_read: SSL_ERROR_SYSCALL, errno 104 This commit uses wget as some have reported success. Closes: #3606 Approved by: davidiw

view details

Ankush Agarwal

commit sha 31ef2f27b63794702a922ff8f753e9a91c617391

[cluster-test] Set Autoscaling Group Size from cluster-test Closes: #3607 Approved by: andll

view details

Junkil Park

commit sha cac17ca0adef35cde3d23935173408bb752f3995

Update script_provider.move Fixed a mysterious string in `script_provider.move` which cause Prover test fail Closes: #3617 Approved by: DavidLDill,

view details

Bob Wilson

commit sha a62fdd025dc6f03e5600dc86f13b787ad6c2efb4

[cli] fix incorrect argument check for compile command This is a follow up to #3425 where I changed the arguments for the compile command. I missed that the minimum number of arguments is now 3 instead of 4. This also changes the help message to refer to "dependency_source_files" with underscores instead of spaces to separate the words. That is more consistent with the other messages. Closes: #3516 Approved by: runtian-zhou

view details

Victor Gao

commit sha ddf372ae26a704694d5c3fb8f3511e2fa4b4f681

[language][vm] non-recursive serializer/deserializer implementation for SignatureToken SignatureToken can be deeply nested, which can cause the serializer and deserializer to overflow the stack. This implements a non-recursive algorithm, using Vec for memory management, eliminating such stack overflows. Closes: #3572 Approved by: dariorussi

view details

Victor Gao

commit sha 40ea7a658c70204c631469b3cb1b151e822d5e49

[language][vm] separate CompiledScript and CompiledModule in the file format This is the beginning of the work to fully seperate scripts and modules. While they share a few common tables for external dependencies, the entry point (main) of a script is still very different from a full fledged function definition in a module. This gets it simplified by replacing the full function definition with only the fields required, also eliminating the need for a few other table entries. Closes: #3500 Approved by: dariorussi

view details

Phoenix Orlov

commit sha e323dce5cac4c804c5c1da61e0f1bdf7949b4579

[mempool] remove lru-cache and ttl-cache dependencies Closes: #3603 Approved by: davidw

view details

Victor Gao

commit sha c1c47b8764ea4a31652963a9fb8845216726824c

[language][vm] non-recursive bounds checker implementation for SignatureToken Reimplement the bounds checker for SignatureToken in a non-recursive fashion to avoid overflowing the stack. Closes: #3596 Approved by: dariorussi

view details

David Wolinsky

commit sha 3d732060398ca0b32bf2749a5fc5013e00b9390e

[genesis] Remove unused private key from API Genesis is an unsigned writeset, it doesn't require a private key as it. Closes: #3608 Approved by: zekun000

view details

David Wolinsky

commit sha 77b018dbf1eb9b5551ae38da7ded11d40a920562

[genesis] Remove NodeConfig from genesis encoding Genesis requires authenticators for accounts in order to create accounts. Passing in NodeConfig is overly specific when all that is needed is the authenticator itself. Closes: #3608 Approved by: zekun000

view details

David Wolinsky

commit sha fbc79fa43eec89c2499caad0a06fc9eaeddc865a

[genesis] Use scripts to initialize ValidatorConfigs Historically, we would generate both ValidatorSet and DiscoverySet to then produce the correct interface into executing functions within the VM. This results in a lot of extra data being generated and an interesting dependency on NodeConfigs. This is a first step toward leveraging the proposed concept for genesis, which would be to leverage signed transactions from the would-be validators on their accounts. This is incremental in that the following need to happen: * Move config::generator into its own crate, eliminate the generation of discovery and validator sets, and generate the scripts there. * Discuss with the move team on the best place for handling TransactionArgument <-> Value * Consider leveraging the LibraVM for verification rather than the MoveVM * All the work that goes into deoupling the automated generation of genesis from the code base and make it map to both. Closes: #3608 Approved by: zekun000

view details

David Wolinsky

commit sha 22e9451bc25153fbcca8146516ab838df80688f7

[language] eliminate a special use-case of config generator Now it accesses a proper ValidatorSet in its own memory. Note: This requires us to have a more proper Ed25519 key due to checks during deserialization... Closes: #3608 Approved by: zekun000

view details

push time in a month

delete branch ankushagarwal/libra

delete branch : cluster-test-autoscale

delete time in a month

pull request commentlibra/libra

[cluster-test] Add a teardown method for ClusterTestRunner

It will leave just the autoscaling group sized up. Subsequent run will not fail, it will correctly set the asg group size during scale up and scale down

ankushagarwal

comment created time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha 0a9a89f83f24c6fe7c2055ab101f968132a3b40d

[cluster-test] Add a teardown method for ClusterTestRunner

view details

push time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha e33fbce67dd8d05babda5093027a8245937119c1

[cluster-test] Add a teardown method for ClusterTestRunner

view details

push time in a month

PR opened libra/libra

[cluster-test] Add a teardown method for ClusterTestRunner

Summary

Add a teardown method for ClusterTestRunner which scales down the cluster to size 1.

Motivation

Now that cluster-test also performs autoscaling, it can conflict with the cluster-autoscaler running on the cluster. This has led to a couple of failures. There is no way to coordinate these two processes, so its easier that cluster-test does both the scaling-up and scaling-down.

Test Plan

Ran single experiment and ci suite

+26 -4

0 comment

2 changed files

pr created time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha 3479708974059aec6bc44e3c12ea71573358c6d4

[cluster-test] Add a teardown method for ClusterTestRunner

view details

push time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha a6da993ed0b351ec4ec99de7565386cffaccbf18

WIP

view details

push time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha c2d9d787e7193be5ad13282d4446baac6f13c72d

WIP

view details

push time in a month

create barnchankushagarwal/libra

branch : cluster-test-scaledown

created branch time in a month

pull request commentlibra/libra

[RFC][Move] Change move build output directory to something less generic

@bors-libra retry

JoelMarcey

comment created time in a month

pull request commentlibra/libra

[cluster-test] Use correct asg name instead of hardcoded value

@bors-libra retry

ankushagarwal

comment created time in a month

PR opened libra/libra

Reviewers
[cluster-test] Use correct asg name instead of hardcoded value cluster_test

Summary

Use correct asg name instead of hardcoded value

+1 -1

0 comment

1 changed file

pr created time in a month

create barnchankushagarwal/libra

branch : cluster-test-autoscale

created branch time in a month

delete branch ankushagarwal/libra

delete branch : cluster-test-size

delete time in a month

push eventankushagarwal/libra

Shaz Qadeer

commit sha a1f9b8211f113b5939bb6e3bc162246c67497d6e

allowed dataflow analysis to compute error output for a block Closes: #3604 Approved by: junkil-park

view details

Wolfgang Grieskamp

commit sha 49cfe3d2fa19bd400b360aac0f6425eb973e4b9f

[move-prover] Refactor test suite driver and copy new lib into test tree. This refactors our test directory and driver as follows: - all top-level tests have been moved into `sources/functional` - a copy of the new standard library has been placed in `sources/new_stdlib` - tests don't need to longer declare their dependencies via the `// dep` comment. Instead, dependencies are automatically detected by using a search path. - different test directories (like functional, stdlib, etc.) can use different flags and search paths. This is configured in `testsuite.rs`, and allows for example `new_stdlib` to find dependencies at a different place than `stdlib`. - for `new_stdlib`, verification is off by default and need to be turned on using `pragma verify = true` (either in a module spec block for the whole module or per function spec block) - a `.exp` file is not longer generated if the test passes without any errors. On UPBL=1, existing .exp files are instead removed if there are no errors. This prepares us to run directly against the standard library source of truth at a later point. There is a nasty issue with native functions of the same name having different signatures in old and new library. To work around this, a hack was added to generate different boogie names so the old/new lib sources can be used by the same prelude. The native functions for the new library, as well as new vector function `index_of`, are not implemented yet and `assert false`. Closes: #3609 Approved by: DavidLDill

view details

Ankush Agarwal

commit sha 769892ed62f122912c208759689ff6dbc3ab17a3

[cluster-test] Set Autoscaling Group Size from cluster-test

view details

push time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha 6bf326bcd3434d8ce94cda0c2b6be24470b07cd1

[cluster-test] Set Autoscaling Group Size from cluster-test

view details

push time in a month

pull request commentlibra/libra

[cluster-test] Set Autoscaling Group Size from cluster-test

So right now, before this PR, autoscaling is managed in the ops repo, but the challenge with that is that we don't get fine grained control. When we launch 60 container, the autoscaler launches exactly 60 nodes.

With this PR, we have more control over what autoscaling looks like and helps us mitigate the AWS issue that we're facing right now

ankushagarwal

comment created time in a month

pull request commentlibra/libra

[cluster-test] Set Autoscaling Group Size from cluster-test

The AWS issue still seems to be happening where we request 63 nodes, only 62 of them get provisioned and 1 is stuck in pending.

Applied calibra/libra-ops#524 to all three clusters and verified that this PR works on all three of them.

ankushagarwal

comment created time in a month

Pull request review commentlibra/libra

[cluster-test] Set Autoscaling Group Size from cluster-test

 impl ClusterUtil {                 "Deploying with {} tag for validators and fullnodes",                 image_tag             );+            let asg_name = format!(+                "{}-k8s-testnet-validators",+                cluster_swarm+                    .get_workspace()+                    .await+                    .expect("Failed to get workspace")+            );+            aws::set_asg_size(+                (args.k8s_num_validators+                    + (args.k8s_fullnodes_per_validator * args.k8s_num_validators))+                    as i64,+                3,

I think CP also uses 30-node config. But either way, percent is a good idea. Updated to use buffer_percent

ankushagarwal

comment created time in a month

Pull request review commentlibra/libra

[cluster-test] Set Autoscaling Group Size from cluster-test

 pod_name=${pod_name/_/-} #underscore not allowed in pod name specfile=$(mktemp) echo "Pod Spec : ${specfile}" join_args $*+ENV="$ENV AWS_ROLE_SESSION_NAME=AWS_ROLE_SESSION_NAME"

So this is intentional. AWS_ROLE_SESSION_NAME needs to be set to some non-empty variable, so I've set it to the literal AWS_ROLE_SESSION_NAME

ankushagarwal

comment created time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha 515e1daf5016ac9596140f4d4e59e698b9a16621

[cluster-test] Set Autoscaling Group Size from cluster-test

view details

push time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha 1b8e4abf838ebaba0abb7c38df21fe8e83cb3946

[cluster-test] Set Autoscaling Group Size from cluster-test

view details

push time in a month

PR opened libra/libra

[cluster-test] Set Autoscaling Group Size from cluster-test

Summary

  • We were seeing issues where an instance was failing to launch in the AWS autoscaling group - causing a libra node to be unassigned image

  • Create set_asg_size which sets the autoscaling group size to the min desired capacity plus an additional buffer

  • When cluster-test starts, set the autoscaling group size to the desired size plus a buffer of 3 nodes to tolerate upto 3 instance launch failures

Test Plan

Ran cluster-test

+234 -11

0 comment

7 changed files

pr created time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha cdf189ab20ecc4454e8f12755f689e0876b00897

[cluster-test] Set Autoscaling Group Size from cluster-test

view details

push time in a month

create barnchankushagarwal/libra

branch : cluster-test-size

created branch time in a month

push eventankushagarwal/libra

dependabot-preview[bot]

commit sha 72791569db71baf189318f02245a7051eafa178c

Bump pin-project from 0.4.8 to 0.4.9 Bumps [pin-project](https://github.com/taiki-e/pin-project) from 0.4.8 to 0.4.9. - [Release notes](https://github.com/taiki-e/pin-project/releases) - [Changelog](https://github.com/taiki-e/pin-project/blob/master/CHANGELOG.md) - [Commits](https://github.com/taiki-e/pin-project/compare/v0.4.8...v0.4.9) Signed-off-by: dependabot-preview[bot] <support@dependabot.com> Closes: #3445 Approved by: wqfish

view details

Young Yang Liauw

commit sha e8be73fff3869c5e6b16c9f90ccb87ec60caed92

[CI] retry E2E tests Retry flaky E2E tests, up to 3 times. Closes: #3352 Approved by: metajack

view details

Young Yang Liauw

commit sha 1609786843190fc32aa9bd9a39795b9d51055a98

[CI] flag flaky E2E tests found Flag any flaky E2E tests found in CI. Closes: #3352 Approved by: metajack

view details

Sam Blackshear

commit sha 8e90e8de5be3777d6d10f6603f9b9c276d80c00f

[move core modules] Compare module for polymorphic comparison of Move values This defines a total order Compare::cmp<T> among Move values of the same type T. That is, we have the guarantee: Compare::cmp<T>(t1, t2) == LESS_THEN iff Compare::cmp<T>(t2, t1) == GREATER_THEN Move has bytecodes for polymorphic equality == and inequality !=. The implementation of Compare::cmp<T>(t1, t2) also guarantees that Compare::cmp<T>(t1, t2) == EQUAL iff t1 == t2 and (similarly) Compare::cmp<T>(t1, t2) != EQUAL iff t1 != t2. We implement Compare.cmp(t1, t2) using the vector<u8>'s produced by LCS::to_bytes(t2) and LCS::to_bytes(t2). We compare the vectors first by length, then byte-by-byte from right to left if the lengths are equal. The length comparison is done in the obvious way; we return LESS_THAN if len(lcs(t1)) < len(lcs(t2)) and GREATER_THAN if len(lcs(t1)) > len(lcs(t2)). This ensures that for all primitive types T with < and > comparison operators exposed in Move bytecode (u8, u64, u128), we have Compare::cmp<T>(t1, t2) == LESS_THEN iff t1 < t2 and (similarly) Compare::cmp<T>(t1, t2) == GREATER_THEN iff t1 > t2. // For all other types, the order is whatever the LCS encoding of the type and the comparison strategy above gives you. One case where the order might be surprising is the address type. Addresses are 16 byte hex values that LCS encodes with the identity function. The right to left, byte-by-byte comparison means that (for example) Compare::cmp(0x01, 0x10) == LESS_THAN (as you'd expect), but Compare::cmp(0x100, 0x001) == LESS_THAN (as you probably wouldn't expect). Keep this in mind when writing code that compares addresses. Finally, a note on the constants EQUAL, LESS_THAN, and GREATER_THAN. Move does not have signed integers, so we cannot use the usual 0, -1, 1 convention for this. Instead, we define a new convention using u8 constants: const EQUAL = 0u8 const LESS_THAN = 1u8 const GREATER_THAN = 2u8 Closes: #3337 Approved by: tnowacki

view details

Bob Wilson

commit sha 9c758e0db3a211c26ea05e31a4f605adfbfdccc0

Use the Move source language for client_proxy and libratest/smoke_test Some changes here may not be obvious: - The compile_program function used to perform an automatic substitution of "{{sender}}" in the code, but I moved that functionality to the caller (e.g., smoke_test.rs). The module.move file needs an "address" directive and that file also needs to be provided as a dependency when compiling script.move. The temporary file with the substituted address was not previously accessible, and I didn't want to blindly rewrite all the dependency files. Doing the substitution in the caller makes it possible to reuse the file. - The source compiler does not support a feature like the IR compiler has to list dependencies. Instead this relies on a new feature in the source compiler (PR #3373) to include the entire Move stdlib as a dependency. - Removing the code to grab the dependency bytecode from on-chain made the get_account_blob function in libra_client.rs unused, so I removed it as well. Closes: #3425 Approved by: tnowacki

view details

Junkil Park

commit sha bb728128263ba54e311bf38cf1e573792e652096

Fixed the builtin spec function $Hash_sha2 and $Hash_sha3 - Fixed the signature of the builtin spec functions $Hash_sha2 and $Hash_sha3 Closes: #3450 Approved by: wrwg

view details

Mathieu Baudet

commit sha c6f2d3bb6adee2ed6769f010bc0747f26a05e1a0

[serde reflection] clean up clowny `mut self`s and unneeded `&mut` Closes: #3393 Approved by: metajack

view details

Mathieu Baudet

commit sha a84f7a1d29808a29cd70904c4036ac3cc6dcfb74

[serde reflection] More precise typing for Deserializer Closes: #3393 Approved by: metajack

view details

Mathieu Baudet

commit sha 753dd28e3945f988667ee831c3a581c621396015

[serde reflection] Supporting borrowing from recorded values Closes: #3393 Approved by: metajack

view details

Mathieu Baudet

commit sha c60ab56b49d952848fd2fdbd3d0a7b3362856723

[LCS] change keys to Bytes to avoid a copy Closes: #3393 Approved by: metajack

view details

Mathieu Baudet

commit sha a93cc3b2e31f225fd8fb90dd86a3c154b928ad2b

[serde reflection] Rename Records into SerializationRecords in an attempt to clarify the new APIs Closes: #3393 Approved by: metajack

view details

Mathieu Baudet

commit sha 190fb0cb04c5d7774942463471e09730b7b3390f

[serde reflection] improve doc and comments Closes: #3393 Approved by: metajack

view details

dependabot-preview[bot]

commit sha 9bf12e8775df03383c358d6ebb2fab26b43763f6

Bump mirai-annotations from 1.6.1 to 1.7.0 Bumps [mirai-annotations](https://github.com/facebookexperimental/MIRAI) from 1.6.1 to 1.7.0. - [Release notes](https://github.com/facebookexperimental/MIRAI/releases) - [Commits](https://github.com/facebookexperimental/MIRAI/commits) Signed-off-by: dependabot-preview[bot] <support@dependabot.com> Closes: #3444 Approved by: rexhoffman

view details

Phoenix Orlov

commit sha adb871253c96847fcdd5028d668f33608629035d

[core types] remove gas_specifier field Closes: #3451 Approved by: tzakian

view details

Todd Nowacki

commit sha 9c325e25efc809df3b4993104d127923af512383

[move vm] Added Constant pool - New pool for constants - The constants can support any non-struct value, serialized in the same format as on chain values - Addresses are split - Runtime constants are in the constant pool - Namespace constants are in the 'AddressIdentifierPool' Closes: #3376 Approved by: dariorussi

view details

Young Yang Liauw

commit sha 88fab0e0455c8c6f6a7871539e4453020b626138

[CI] add missing comment in e2e test retry Add the missing comment in the e2e test retry logic. Also add 10 sec between retries. Closes: #3452 Approved by: rexhoffman

view details

Runtian Zhou

commit sha ebcd6674f7afb0b15b0db5230c8620dd64e1408e

[vm] Load the writeset from StateView before committing it to the executor Closes: #3415 Approved by: msmouse

view details

Victor Gao

commit sha d80e7daeec744c2e4a179c953230b1e04e4464c1

[language] ir tests revert to staged genesis Closes: #3455 Approved by: tnowacki

view details

Aaron Gao

commit sha 7ab9566c3a838c8127960ca7dfd6d9f93d12cbaf

[executor] split executor public methods into two traits Closes: #3459 Approved by: lightmark

view details

aldenhu

commit sha dbdec7228af94bcb1c26336f5c58b91fc3e049bc

[storage] move VerifiedStateView out of storage-client Closes: #3461 Approved by: lightmark

view details

push time in a month

push eventankushagarwal/libra

aldenhu

commit sha a3c13337356204bd3c99dea9225b3af64c4770eb

[storage] TreeState be able to represent empty transaction accumulator (w/wo pre-genesis state) Closes: #3304 Approved by: wqfish

view details

aldenhu

commit sha 4f0286b47bc5d2612ea9e4600524b3dd04b4e300

[storage] get_latest_tree_state Closes: #3304 Approved by: wqfish

view details

David Wolinsky

commit sha bce58d52a3a8ba2f2a84aea83195c0385a8d8a32

[secure-storage] Rename some methods to better match their intents We don't get private keys, we export them -- this matches the Vault API, which is more explicit than the get API -- as this will be a different permission. We don't generate private keys, we create them -- this matches the KV API, generate didn't match any existing APIs. Closes: #3276 Approved by: joshlind

view details

David Wolinsky

commit sha 5c33e4a08b8d375c963456bdd24dcc688430a6d5

[secure-storage] Use Vault Transit * This uses Vault transit instead of secrets to handle crypto * The transit doesn't handle namespaces like secrets. It cannot handle paths, so namespaces are postfixed with __, e.g., namespace__key_name. * Transit does not support whitespace in names (Test Key) so all (Test Key) have been replaced by (Test_Key) * This doesn't implement trim, so it is a bit inefficient in retrieving keys, this shouldn't be an issue immediately and might be able to be configured by policy * I tried to avoid making too many non-critical changes in this commit, will follow up with that in the next... * This does not implement policies properly for keys... it might but it isn't tested. Closes: #3276 Approved by: joshlind

view details

David Wolinsky

commit sha 4ef05381edd2f53efbdaceb47222ac5b987f0551

[secure-storage] vault crypto capabilities The previous commit ignored policies for Vault's new crypto API. This was done to keep the commit somewhat accessible. In this commit, we address that: * Several new Libra capabilities to map to the appropriate crypto action * Individualized tests to show how those map to vault * A few minor bugs were addressed during testing * Introduced a bit more flexibility into the vault API -- to be honest, I'm not really happy with the internals of either the libra vault client or the vault storage, but I suspect it might be better to investigate other storage solutions before coming back to fix this one. Closes: #3276 Approved by: joshlind

view details

David Wolinsky

commit sha 53de44871f14a276bc2cec00bd80857a82d1dd0c

[secure-storage] response -> resp to reduce lines In an effort to keep the code tighter and more readable, renaming these variables. Closes: #3276 Approved by: joshlind

view details

David Wolinsky

commit sha 0c100e30630a3de9f98ca99e38244e734c374383

[secure-storage] Add ability to reset vault policies Not being able to reset policies doesn't really ensure you have proper test coverage... Closes: #3276 Approved by: joshlind

view details

David Wolinsky

commit sha 3f0d876b5d61b123d5cc6a1159bda5a340e8fc71

[vault] use TIMEOUT const instead of 10_000 for request timeout Closes: #3276 Approved by: joshlind

view details

David Wolinsky

commit sha 255503e94406f49a70f69e8cb4f0c6fa5d03ebf0

[docker] remove redundant named dockerfiles Doing this in a separate commit so we can incrementally adjust libra circle ci to handle the name change. Closes: #3279 Approved by: sausagee

view details

aldenhu

commit sha 3b0107a8640c7c0956a16da1696692bb55945b2b

[storage] VerifiedStateView depends on DbReader trait instead of StorageRead Closes: #3310 Approved by: wqfish

view details

aldenhu

commit sha aa8404661225ca436d395ed9392dd1a2e438143a

[executor] depend on DbReader instead of StorageRead Closes: #3311 Approved by: wqfish

view details

Qinfan Wu

commit sha 9ed252cffcf34e6f0f8a6d095607dd60f5cbf767

[Storage] Move rand to dev dependency Closes: #3293 Approved by: msmouse

view details

Aaron Gao

commit sha c7d39d8b79978b5070c8285f5ab6925354f13203

[storage-if] create storage-service related types Closes: #3314 Approved by: msmouse

view details

dependabot-preview[bot]

commit sha 920b263b26c0322badc0c7984f063fdee3bf42ff

Bump async-trait from 0.1.29 to 0.1.30 Bumps [async-trait](https://github.com/dtolnay/async-trait) from 0.1.29 to 0.1.30. - [Release notes](https://github.com/dtolnay/async-trait/releases) - [Commits](https://github.com/dtolnay/async-trait/compare/0.1.29...0.1.30) Signed-off-by: dependabot-preview[bot] <support@dependabot.com> Closes: #3312 Approved by: rexhoffman

view details

Wolfgang Grieskamp

commit sha 9640cd89e9eaa979c804c132f1beb596a45fe41a

[move-prover] Prevent aliasing of top-level reference parameters. This implements a simpler regime for ensuring exclusiveness of top-level reference parameters (those which are passed into a top-level `foo_verify` boogie procedure). For each such parameter, we ensure that the Reference is disjoint from any other referenc, based on the following properties: - A mutable reference is by Move borrow semantics always exclusive from every other reference. - An immutable reference behaves like a value and that it is a reference is irrelevant, so we can choose an arbitray reference for it (we happen to chose one which is disjoint from others) Previously, we had a more complicated mechanism to deal with reference aliasing based on the stratified function `$IsValidReferenceParameter`, which is removed in this PR. The previous mechanism was also only preventing aliasing for locals on the stack, not for globals. The new implementation simply defines a new location type `Param(i: int): Location` which we assign to top-level references, where `i` is the parameter position. We generate code as such: ``` procedure foo_verify(ref: Reference) { assume l#Reference(ref) == Param(0); assume size#Path(p#Reference(ref)) == 0; call foo(ref); } ``` All existing tests pass, and a new test has been added for the case of aliasing global mut refs. Closes: #3307 Approved by: shazqadeer

view details

Wolfgang Grieskamp

commit sha 089092a6cc6d73dbb27e211ed6da1d2847f8f919

[move-prover] Fixing a bug in update invariants in presence of branching. Until now, we had wrongly assumed that a mut ref can only have exactly one before-borrow snapshot associated. However, due to branching this is not the case. Consider e.g.: ``` let r: &mut R; if (b) r = &mut t1 else r = &mut t2; ``` Here `r` can represent either `&mut t1` or `&mut t2`, depending on the value of `b`. In this PR, we extend the tracking of before-update index from a single value per reference to a set of indices. In the example above, the reference `r` will have assigned both the before-borrow indices of `&mut t1` and `&mut t2`. In order to decide which invariant has to be enforced when `r` dies, we use a new boolean variable `$before_borrow_N_used` which is managed at runtime and set to `true` once the associated borrow instruction is executed. This also fixes a bunch of related issues, like incorrectly asserting invariants for borrows which are never executed because of branching. There are likely more issues with this if it comes to loops, and we eventually will need to move the current ad-hoc analysis we do on the bytecode to the DFA framework and peform it in a better principled way. Closes: #3280 Approved by: shazqadeer

view details

David Dill

commit sha c0d1794e5ef0b887589f068c7c7deb3cbe9b33a0

[move-prover] Added Boogie model for serialize to prelude.bpl Move datastructures are converted to vector<u8> for sha2/3, crypto and other purposes by a native library called "serialize". There is no "deserialize" in Move and no plans to create one. The new Boogie code for serialize maps a Move value to an arbitrary non-empty vector<8>. There is a an additional axiom that constrains serialize to be an injection, because (I believe) the correctness of Move modules will sometimes depend on this property. However, the model is a conservative approximation of the actual function because it does not guarantee that the bytes in the serialized value correspond to what the native serialize function does, or even that the length of the result is the same as the actual serialize. We should be able to verify properties that require serialize to be an injection, but won't be able to verify code that accesses and manipulates bytes from a serialized value or, for example, code that depends on the length of serialized values. There are test cases serialize-model.move and serialize-model-invalid.move, also. Some new .exp files were created for stdlib files that use lcs.move. Closes: #3309 Approved by: wrwg

view details

Brandon Williams

commit sha b60264d77892a7be8c2b06b20f2581368df97601

datatest: enable listing of tests Teach test binaries built using `datatest-stable` to be able to list the names of each test with the `--list` option. This matches the behavior of the `--list` option with test binaries built using the default test harness. Closes: #3318 Approved by: sausagee

view details

Runtian Zhou

commit sha ec9711d69ae7d5b60c1f3659e7948aa57179a2f9

[vm] Refactor transaction verification logic Closes: #3299 Approved by: sblackshear

view details

Runtian Zhou

commit sha ffa6f1d8fe908da69d153e6f8f8f868c632227e0

[move] Implement the authentication rules for TransactionPayload::WriteSet Closes: #3299 Approved by: sblackshear

view details

push time in a month

delete branch ankushagarwal/libra

delete branch : retry-kubectl

delete time in a month

pull request commentlibra/libra

Use retries when building docker image

@bors-libra r=sausagee

ankushagarwal

comment created time in a month

push eventankushagarwal/libra

dependabot-preview[bot]

commit sha 654a6418ad0eeba2bccc92412cac3f0bdcc08fb9

Bump kube from 0.31.0 to 0.32.0 Bumps [kube](https://github.com/clux/kube-rs) from 0.31.0 to 0.32.0. - [Release notes](https://github.com/clux/kube-rs/releases) - [Changelog](https://github.com/clux/kube-rs/blob/master/CHANGELOG.md) - [Commits](https://github.com/clux/kube-rs/compare/0.31.0...0.32.0) Signed-off-by: dependabot-preview[bot] <support@dependabot.com> Closes: #3431 Approved by: ankushagarwal

view details

Rex Hoffman

commit sha a1adc8829b59235bea35dcafaa070b4309a14ce2

update to the ever changing kube api Closes: #3431 Approved by: ankushagarwal

view details

Junkil Park

commit sha f4bb5cde04185784c456d1de77c35857ff8b61cf

Verified prologue in LibraAccount - Created and verified a modified version of `prologue` avoiding the multiplication issue. - Added the native spec functions in the Hash module, and modified Prelude accordingly - Specified and verified a few other functions in LibraAccount - withdraw_from_sender - withdraw_with_capability Closes: #3432 Approved by: DavidLDill

view details

Qinfan Wu

commit sha ccbc3d432f29d677179c72db1bbc4de39acda465

[Executor] Clean up a bit dependencies Closes: #3437 Approved by: lightmark

view details

Abhay Bothra

commit sha 97b7eda88acbd8d6052b32cf250dfad83145b491

[network] Add ErrorCode::NotSupported to wire protocol ErrorCode::ParsingError does not capture the case where we support a protocol in the code, but don't want to expose it to a client. An example might be the internal protocols like the one that will be created for Validator -> SafetyRules communication - these protocols can be correctly parse, but should not be exposed to external nodes. Closes: #3436 Approved by: metajack

view details

Joshua Lind

commit sha 910800570de5bc784d9d2852a8132090bccf3307

[Secure JSON RPC] Implement a minimal JSON RPC client for security-critical components. Closes: #3325 Approved by: davidiw

view details

Christina Lee

commit sha 840811f5f45dc5bb90e33f4f9707a6fbafcea412

[mempool] create shared_mempool folder in crate Closes: #3380 Approved by: phoenix-antigravity

view details

Christina Lee

commit sha c2cbc5aa72e980ed6eaae3fe1e8fa1d2da11597d

[mempool] make tests top folder in crate Closes: #3380 Approved by: phoenix-antigravity

view details

Clark Barrett

commit sha 0d71cec16a57d39dce7e344d91702c22e016ffcd

Clean up of prelude.bpl: - fixed a couple of typos in comments - Removed redundant ValueArray function (AddalueArray same as ExtendValueArray) - Many instructions were using "assume" for the types of their input parameters - these should be enforced by requires - fixed this - Changing the above uncovered a couple of issues - Unimplemented instructions lose type information - fixed for now by adding a dummy call that preserves typing information - Vector function models that were using loops broke requirements because of loop havocing - I commented these out for now Closes: #3389 Approved by: cbarrettfb

view details

Ankush Agarwal

commit sha 78d38b4ab8474fc434c52cea33d9598a596183b2

Add retry logic for docker build scripts

view details

push time in a month

pull request commentlibra/libra

Use retries when building docker image

@bors-libra r=sausagee

ankushagarwal

comment created time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha 65d5ab96ef6380e27c5a8522774fd4795608e581

Add retry logic for docker build scripts

view details

push time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha 92be56178c558864347623b2c7b74c08686ef282

Add retry logic for docker build scripts

view details

push time in a month

pull request commentlibra/libra

Use retries when building docker image

@bors-libra r=sausagee

ankushagarwal

comment created time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha 607d7e949cb413a2fedaeaa75ac3fd1b555ea1ae

Add retry logic for docker build scripts

view details

push time in a month

pull request commentlibra/libra

Use retries when building docker image

@bors-libra r=sausagee

ankushagarwal

comment created time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha c8e46b8514a66b5c54e744ca96e0ebc172dbf32b

Add retry logic for docker build scripts

view details

push time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha fecaa23712378706f448d0d769da29cb69642b26

Add retry logic for docker build scripts

view details

push time in a month

Pull request review commentlibra/libra

Use retries when building docker image

 RUN rustup install $(cat rust-toolchain) COPY . /libra RUN docker/cluster-test/compile.sh FROM debian:buster-RUN apt-get update && apt-get install -y openssh-client curl && curl -L "https://storage.googleapis.com/kubernetes-release/release/`curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt`/bin/linux/amd64/kubectl" -o /usr/local/bin/kubectl && chmod +x /usr/local/bin/kubectl

Done

ankushagarwal

comment created time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha 451fb83b4f92e1cee535f3e0ad06dd4b64a73f6c

Add retry logic for docker build scripts

view details

push time in a month

push eventankushagarwal/libra

Wolfgang Grieskamp

commit sha 1650761600770c0dd15a75ee96e7e564ab121284

[move-prover] Streamlining bytecode, plus make spec blocks instructions. This further streamlines and minimizes the stackless bytecode representation: - Collapses all instructions similar to a call (user function, pack, unpack, borrow, exists, freeze, unary&binary op, etc) into one single instruction with a nested enum. This reduces the number of top-level instructions down to 8, making traversals simpler. - Removes the `Labeled(attr_id, label, bytecode)` instruction and replaces it with `Label(attr_id, label)`, making labels indepdendent instructions. This makes it easier to transform bytecode, and avoids implicit constraints on the instruction type, like that Labeled instructions can't be nested. - Introduces spec blocks as instructions. Both spec blocks given by the user as those which may be generated by tramsformations are supported. Closes: #3426 Approved by: shazqadeer

view details

Rain

commit sha 0aaead90f7259eaa80a2131e6f417071328d7e10

[x] upgrade to guppy 0.3.0 The new APIs suggested by @metajack mean the workspace hack linter is simpler. Closes: #3428 Approved by: metajack

view details

Ankush Agarwal

commit sha ee14dd5817637f9e538366fcceefdb2939c0715b

[k8s][cluster-test] Delete old jobs in delete_all Closes: #3427 Approved by: andll

view details

Qinfan Wu

commit sha 333fe5daa4df02e703bb98b228bc9c365a47dd8a

[Types] Get rid of path hack Closes: #3433 Approved by: msmouse

view details

Qinfan Wu

commit sha ac696408df372db10e63e0990daf75a15ef98eeb

[Crypto] Get rid of path hack Closes: #3433 Approved by: msmouse

view details

Bob Wilson

commit sha e1f58a5282eef0fbe65ddf6fcda0626aa6172de2

[cli] add consistency assertions for client_proxy commands When the client_proxy commands are dispatched through the DevCommand struct, the first argument determines which command is executed, but when the commands are invoked directly, that first argument is ignored. I noticed several cases where the command argument was incorrect and was confused for a while. This adds assertions to check that the command argument matches the command being executed and fixes the places that were inconsistent. Closes: #3421 Approved by: christinacelee

view details

Bob Wilson

commit sha cf0d090219c8788d1b305dd0b077574e0287c873

[move-lang] allow specifying directories for input files and dependencies When compiling a Move file that is not part of the standard library, there are typically a number of dependencies on the stdlib modules. It is a hassle to specify these individually. This change lets you specify the directory containing the stdlib modules as a dependency, and the compiler will expand that to include all the ".move" files inside that directory. We can do the same for the input files to make it easier to build the stdlib (or any other collection of Move files). This also changes the move-build and move-check programs to customize the error printing. The default output for io::Error results is not very pretty. Closes: #3373 Approved by: tnowacki

view details

Ankush Agarwal

commit sha 6c3bab40dc0ac18445470367d78253c9997a3b15

Add retry logic for docker build scripts

view details

push time in a month

Pull request review commentlibra/libra

Use retries when downloading kubectl

 RUN rustup install $(cat rust-toolchain) COPY . /libra RUN docker/cluster-test/compile.sh FROM debian:buster-RUN apt-get update && apt-get install -y openssh-client curl && curl -L "https://storage.googleapis.com/kubernetes-release/release/`curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt`/bin/linux/amd64/kubectl" -o /usr/local/bin/kubectl && chmod +x /usr/local/bin/kubectl+RUN apt-get update && apt-get install -y openssh-client curl+# Use retries to download kubectl+RUN ["bash", "-c", "download_kubectl() {   local location=/usr/local/bin/kubectl;   for ((i = 0; i < 30; i++)); do     curl -L \"https://storage.googleapis.com/kubernetes-release/release/`curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt`/bin/linux/amd64/kubectl\" -o ${location};     if [ $? -gt 0 ]; then       echo \"curl failed. retrying in 10 secs.\";       sleep 10;     else       chmod +x ${location};       echo \"successfully installed kubectl\";       return 0;     fi;   done;   return 1; }; download_kubectl"]

Another thought, can we leverage DLC to avoid re-downloading kubectl altogether

We are already leveraging DLC. This is using the builder pattern by building libra in a separate container. Its confusing because there is no newline b/w the two builds. I have added one.

ankushagarwal

comment created time in a month

Pull request review commentlibra/libra

Use retries when downloading kubectl

 RUN rustup install $(cat rust-toolchain) COPY . /libra RUN docker/cluster-test/compile.sh FROM debian:buster-RUN apt-get update && apt-get install -y openssh-client curl && curl -L "https://storage.googleapis.com/kubernetes-release/release/`curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt`/bin/linux/amd64/kubectl" -o /usr/local/bin/kubectl && chmod +x /usr/local/bin/kubectl+RUN apt-get update && apt-get install -y openssh-client curl+# Use retries to download kubectl+RUN ["bash", "-c", "download_kubectl() {   local location=/usr/local/bin/kubectl;   for ((i = 0; i < 30; i++)); do     curl -L \"https://storage.googleapis.com/kubernetes-release/release/`curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt`/bin/linux/amd64/kubectl\" -o ${location};     if [ $? -gt 0 ]; then       echo \"curl failed. retrying in 10 secs.\";       sleep 10;     else       chmod +x ${location};       echo \"successfully installed kubectl\";       return 0;     fi;   done;   return 1; }; download_kubectl"]

Makes sense. Will do.

ankushagarwal

comment created time in a month

Pull request review commentlibra/libra

Use retries when downloading kubectl

 RUN rustup install $(cat rust-toolchain) COPY . /libra RUN docker/cluster-test/compile.sh FROM debian:buster-RUN apt-get update && apt-get install -y openssh-client curl && curl -L "https://storage.googleapis.com/kubernetes-release/release/`curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt`/bin/linux/amd64/kubectl" -o /usr/local/bin/kubectl && chmod +x /usr/local/bin/kubectl+RUN apt-get update && apt-get install -y openssh-client curl

Looks like thats in a separate build image

ankushagarwal

comment created time in a month

delete branch ankushagarwal/libra

delete branch : cleanup-jobs-2

delete time in a month

pull request commentlibra/libra

[deps] Kube 0.32.0

@bors-libra r+

rexhoffman

comment created time in a month

pull request commentlibra/libra

[k8s][cluster-test] Delete old jobs in delete_all

@bors-libra retry

ankushagarwal

comment created time in a month

push eventankushagarwal/libra

Ankush Agarwal

commit sha 9e47024c1247e8c5b497ca3ba6191dd78d5c8fb6

Use retries when downloading kubectl

view details

push time in a month

PR opened libra/libra

Reviewers
Use retries when downloading kubectl

Summary

  • Failing to download kubectl was causing CI failures.
  • This adds a retry mechanism when downloading kubectl
download_kubectl() {
  local location=/tmp/kubectl;
  for ((i = 0; i < 30; i++)); do
    curl -L "https://storage.googleapis.com/kubernetes-release/release/`curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt`/bin/linux/amd64/kubectl" -o ${location};
    if [[ $? -gt 0 ]]; then
      echo "curl failed. retrying in 10 secs.";
      sleep 10;
    else
      chmod +x ${location};
      echo "successfully installed kubectl";
      return 0;
    fi;
  done;
  return 1;
}

Test Plan

Verified that the RUN commands correctly downloads kubectl with retries

+12 -4

0 comment

4 changed files

pr created time in a month

create barnchankushagarwal/libra

branch : retry-kubectl

created branch time in a month

pull request commentlibra/libra

[k8s][cluster-test] Delete old jobs in delete_all

@bors-libra r=andll

ankushagarwal

comment created time in a month

delete branch ankushagarwal/libra

delete branch : cleanup-jobs

delete time in a month

more