-
Notifications
You must be signed in to change notification settings - Fork 3.8k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
roachtest: cdc/crdb-chaos/rangefeed=true failed [skipped] #37716
Comments
SHA: https://github.com/cockroachdb/cockroach/commits/630a6e9cb3771912cd138f9aa3bea1f0ca9fa7c9 Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1306250&tab=buildLog
|
That's close enough that it may not indicate that anything went seriously wrong. @danhhz can I assign this to you to make a decision on what to do? |
SHA: https://github.com/cockroachdb/cockroach/commits/db98d5fb943e0a45b3878bdf042838408e9aee40 Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1308281&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/fc7e48295cd05f94fd2883498d96d91ad538e559 Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1308263&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/c280de40c2bcab93c41fe82bef8353a5ecd95ac4 Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1311970&tab=buildLog
|
This is related to #36879. The threshold really should be more like 2-3m but we don't have predictable behavior around crdb chaos. The previous failures seemed to all be between 11-12m so I was hoping there was a 10m timeout somewhere that bumping to 15 would fix. At this point, I think we need to investigate what's going on here. This actually would be a good starter/intermediate issue for someone. Possibly me? Could also be you or tobi to try to spread the changefeed debugging skills, but last time I looked into it, it seemed like closed timestamps taking a while to unstick after node chaos, so dunno if it's really a "changefeed" issue |
SHA: https://github.com/cockroachdb/cockroach/commits/61715f0f96f519d599eec6541bbee7394d63209a Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1312952&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/8f42e0d9948256af8b3e1994d514314ba1718c48 Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1315162&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/83e62d69214aaa0f7b976f764b97b0e21a41cde3 Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1318703&tab=buildLog
|
Unassigning myself so someone else can get some cdc exposure. |
SHA: https://github.com/cockroachdb/cockroach/commits/5a88de2233e1405c0553f2d5380fd24218fac3d2 Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1324169&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/8e7ef35a8e4169ec63dc5a4df963d8b31a3d5b61 Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1324151&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/8892e379d84a36b29003420189edd1e10db41d71 Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1328407&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/db6d4425d65bdb027624972ccb19d7aad0bc57cc Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1339372&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/0854bf6d9dd30b4893c19a6c0c3a08809c3748c8 Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1351925&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/da56c792e968574b8f1d9ef3fdb45d56a530221a Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1415578&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/98d6832e9f9edb7e554aaa90d9d4296bb00af16e Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1433695&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/8ebcdac113118ae5fbcaddeecd269f59399aea8c Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1443904&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/e8faca611a902766154ed82581d6d3a7483ad231 Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1462518&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/96b1500e20575ee5c609a00857c78c918078c99b Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1465459&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/4784fe3c51545db5fb5d411937ec1db2ef2b9761 Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1472753&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/62b1678f652461bbc1aaf6bc2c0dd03105ce0ebe Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1488785&tab=buildLog
|
SHA: https://github.com/cockroachdb/cockroach/commits/c6342c90a7fa4ceb1b674faa47a95e1726d05e79 Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1496387&tab=artifacts#/cdc/crdb-chaos/rangefeed=true
|
SHA: https://github.com/cockroachdb/cockroach/commits/1ed03c9811f01fef31950a5cb75a7b591af6fc26 Parameters: To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1518416&tab=artifacts#/cdc/crdb-chaos/rangefeed=true
|
Okay, so this max latency failure is definitely interesting and worth looking into, but not now. It does seem to always get itself started back up, just taking much longer than we'd expect (this assertion really should be around 2m, but we've upped it to try and reduce flakes). I don't think we get a lot of value from running this every night, especially while cdc is not under active development. I'm going to skip this in the interest of reducing noise. |
(Fixed the issue name - I think putting |
Whoops, for some reason I specifically remember you telling me that it worked as long as it was contained in the title (not just a prefix). Anyway, thanks for fixing it! Is this written down anywhere? |
No, and I'm not even aware that anyone else does it (but I like it). It
does work for the issue poster as long as it's somewhere in the title, but
we also have roachdash and other tools that parse the title, and they may
or may not get confused by mangling the prefix.
|
@danhhz is there anything we should do here for the upcoming release? |
With infinite bandwidth, yes, I'd still love for someone to dig into this. Practically... no. |
Discussed in the team meeting but this could've been resolved by us having addressed #48553. Given we found it difficult to repro, I'd be happy to unskip it going forward. |
54201: roachtest: unskip cdc/crdb-chaos r=aayushshah15 a=aayushshah15 I ran this test a total of 15 times in parallel and wasn't able to reproduce. Since its been skipped for 2+ releases, it's hard to know what fixed it, but a good guess is #48561. Release note: None Fixes #37716 Informs #36879 Release justification: testing only 57170: util/log: new experimental integration with Fluentd r=itsbilal a=knz Release note (cli change): It is now possible to redirect logging to [Fluentd](https://www.fluentd.org)-compatible network collectors. See the documentation for details. This is an alpha-quality feature. 59741: opt: fix panic in GenerateLookupJoin r=mgartner a=mgartner #### opt: fix panic in GenerateLookupJoin In #57690 a new code path was introduced from `findConstantFilterCols` from `GenerateLookupJoins`. This new code path made it possible for the filters passed to `findConstantFilterCols` to contain columns that are not part of the given table. This violated the assumption that the filter only contains columns in the given table and caused a panic. This commit fixes the issue by neglecting constant filters for columns not in the given table. Fixes #59738 Release note (bug fix): A bug has been fixed that caused errors when joining two tables when one of the tables had a computed column. This bug was present since version 21.1.0-alpha.2 and not present in any production releases. #### opt: move findConstantFilterCols to general_funcs.go Release note: None 59779: flowinfra: deflake a test r=yuzefovich a=yuzefovich Previously, a unit test could fail in rare circumstances when relocating a range to a remote node, and now we will use SucceedsSoon to avoid that. Also unskip the vectorized option. Fixes: #59712 Release note: None Co-authored-by: Aayush Shah <aayush.shah15@gmail.com> Co-authored-by: Raphael 'kena' Poss <knz@thaumogen.net> Co-authored-by: Marcus Gartner <marcus@cockroachlabs.com> Co-authored-by: Yahor Yuzefovich <yahor@cockroachlabs.com>
SHA: https://github.com/cockroachdb/cockroach/commits/7009f8750d5c3af32d5c43011869048ea7a311ae
Parameters:
To repro, try:
Failed test: https://teamcity.cockroachdb.com/viewLog.html?buildId=1300930&tab=buildLog
The text was updated successfully, but these errors were encountered: