Uh oh!
There was an error while loading. Please reload this page.
HBASE-26807 Unify CallQueueTooBigException special pause with CallDroppedException - #4273
Conversation
Apache-HBase
commented
Mar 24, 2022
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Mar 24, 2022
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Mar 24, 2022
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Mar 25, 2022
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Mar 25, 2022
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Mar 25, 2022
💔 -1 overall
This message was automatically generated. |
c42ca47 to
56cccb4CompareApache-HBase
commented
Mar 25, 2022
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Mar 25, 2022
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Mar 25, 2022
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Mar 25, 2022
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Mar 25, 2022
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Mar 26, 2022
💔 -1 overall
This message was automatically generated. |
56cccb4 to
f1eb85fCompareApache-HBase
commented
Mar 31, 2022
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Mar 31, 2022
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Apr 1, 2022
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Apr 4, 2022
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Apr 4, 2022
💔 -1 overall
This message was automatically generated. |
Apache-HBase
commented
Apr 4, 2022
💔 -1 overall
This message was automatically generated. |
| private final int maxKeyValueSize; | ||
| AsyncConnectionConfiguration(Configuration conf) { | ||
| ConnectionConfiguration connectionConf = new ConnectionConfiguration(conf); |
There was a problem hiding this comment.
note this difference between master and branch-2 -- since we must maintain both an async and blocking configuration, I pulled this in here to DRY up the handling (and remove the duplicate logging)
| public SingleRequestCallerBuilder<T> pauseForCQTBE(long pause, TimeUnit unit) { | ||
| this.pauseForCQTBENs = unit.toNanos(pause); | ||
| public SingleRequestCallerBuilder<T> pauseForServerOverloaded(long pause, TimeUnit unit) { |
There was a problem hiding this comment.
this class is IA.Private, so I simply rename all of these methods rather than handle deprecation
bbeaudreault
commented
Apr 7, 2022
@ndimiduk We should let this branch run the pre-commit checks before merging, but I just checked and it should apply cleanly to branch-2.5 as well. |
Apache-HBase
commented
Apr 7, 2022
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Apr 7, 2022
💔 -1 overall
This message was automatically generated. |
bbeaudreault
commented
Apr 7, 2022
Ok unfortunately I think the test failure might be real. I was able to reproduce locally, and branch-2 succeeded. I don't have time to look into this right now because I'm headed out of town. I'll be back to look at it on Tuesday, so we might want to hold off merging until I can get to it @ndimiduk |
Apache-HBase
commented
Apr 7, 2022
💔 -1 overall
This message was automatically generated. |
…for server overloaded to blocking client
bfb330c to
c3a2a55Comparebbeaudreault
commented
Apr 12, 2022
The test failure was due to a somewhat implicit dependency between HTableMultiplexer and AsyncProcess conf parsing. I fixed that by creating a new AsyncProcess constuctor for use by HTableMultiplier which upgrades the dependency to explicit. This way we're unlikely to hit issues like this again in the future. The change is covered by existing tests since they failed previously and now succeed. |
Apache-HBase
commented
Apr 12, 2022
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Apr 13, 2022
🎊 +1 overall
This message was automatically generated. |
Apache-HBase
commented
Apr 13, 2022
🎊 +1 overall
This message was automatically generated. |
ndimiduk
commented
Apr 19, 2022
Your extra commit for branch-2 looks fine to me. Will you need to apply that change as an addendum to |
bbeaudreault
commented
Apr 19, 2022
I don't think we need an addendum for master. All of the extra commits here are just to support branch-2 specific functionality (i.e blocking client). HTableMultiplexer doesn't exist in master. The one divergence between master and branch-2 that might matter is the difference in in AsyncConnectionConfiguration, where here we instantiate a |
bbeaudreault
commented
Apr 19, 2022
I was thinking of creating a jira to clean up ConnectionConfiguration in master, but not sure how much time I'll have for that in the near future. |
…ppedException (#4273) Signed-off-by: Nick Dimiduk <ndimiduk@apache.org>
…pause with CallDroppedException (apache#4273) Signed-off-by: Nick Dimiduk <ndimiduk@apache.org>
This is the branch-2 port for #4180
The first commit is a simple cherry-pick of that PR.
The second commit applies the same changes to the blocking RPC client classes. In implementing for blocking client, I consolidated the configuration parsing within ConnectionConfiguration. In my opinion this is much cleaner/DRY, and also mirrors how AsyncConnectionConfiguration works for the async client.
Added tests for each of the 3 places that these pauses end up in the blocking client: AsyncProcess, RpcRetryingCallerImpl, and ConnectionImplementation.locationRegions.