[GitHub] nicmichael commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads.

nicmichael commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads. URL: https://github.com/apache/bookkeeper/pull/1804#issuecomment-438063537 Sounds good to me. The most common case is probably addressed by the earlier change.

[GitHub] sijie commented on a change in pull request #1808: Allow to configure sticky reads

sijie commented on a change in pull request #1808: Allow to configure sticky reads URL: https://github.com/apache/bookkeeper/pull/1808#discussion_r232909063 ## File path: bookkeeper-server/src/main/java/org/apache/bookkeeper/client/PendingReadOp.java ## @@ -96,13 +96,38

[GitHub] merlimat opened a new pull request #1808: Allow to configure sticky reads

merlimat opened a new pull request #1808: Allow to configure sticky reads URL: https://github.com/apache/bookkeeper/pull/1808 ### Motivation Currently the BK client is issuing the read requests in round-robin fashion across all the bookies in the write set. One issue with

[GitHub] eolivelli commented on a change in pull request #1808: Allow to configure sticky reads

eolivelli commented on a change in pull request #1808: Allow to configure sticky reads URL: https://github.com/apache/bookkeeper/pull/1808#discussion_r232907899 ## File path: bookkeeper-server/src/main/java/org/apache/bookkeeper/client/PendingReadOp.java ## @@ -96,13

[GitHub] eolivelli commented on issue #1808: Allow to configure sticky reads

eolivelli commented on issue #1808: Allow to configure sticky reads URL: https://github.com/apache/bookkeeper/pull/1808#issuecomment-438150170 We also need tests. Is this a WIP ? This is an automated message from the

[GitHub] ivankelly commented on issue #1641: Added module to enable CPU affinity

ivankelly commented on issue #1641: Added module to enable CPU affinity URL: https://github.com/apache/bookkeeper/pull/1641#issuecomment-437813830 @merlimat thanks. Do you have any numbers for just this change? Like without the spin-wait stuff and HT disablement?

[GitHub] sijie commented on issue #1800: Issue 1799: bkctl is broken with default build options

sijie commented on issue #1800: Issue 1799: bkctl is broken with default build options URL: https://github.com/apache/bookkeeper/pull/1800#issuecomment-437832217 > Cookie should probably be under infrastructure, but that's not for this PR. Cookie change and grouping change are done

[GitHub] ivankelly commented on a change in pull request #1800: Issue 1799: bkctl is broken with default build options

ivankelly commented on a change in pull request #1800: Issue 1799: bkctl is broken with default build options URL: https://github.com/apache/bookkeeper/pull/1800#discussion_r232598241 ## File path: site/docs/latest/getting-started/installation.md ## @@ -60,7 +66,9 @@

[GitHub] ivankelly commented on a change in pull request #1800: Issue 1799: bkctl is broken with default build options

ivankelly commented on a change in pull request #1800: Issue 1799: bkctl is broken with default build options URL: https://github.com/apache/bookkeeper/pull/1800#discussion_r232603952 ## File path: site/docs/latest/getting-started/installation.md ## @@ -60,7 +66,9 @@

[GitHub] ivankelly commented on a change in pull request #1787: [stats] introduce `StatsDoc` annotation for better documenting metrics exposed by bookkeeper

ivankelly commented on a change in pull request #1787: [stats] introduce `StatsDoc` annotation for better documenting metrics exposed by bookkeeper URL: https://github.com/apache/bookkeeper/pull/1787#discussion_r232591697 ## File path: dev/stats-doc-gen ## @@ -0,0 +1,65

[GitHub] eolivelli commented on a change in pull request #1800: Issue 1799: bkctl is broken with default build options

eolivelli commented on a change in pull request #1800: Issue 1799: bkctl is broken with default build options URL: https://github.com/apache/bookkeeper/pull/1800#discussion_r232601780 ## File path: site/docs/latest/getting-started/installation.md ## @@ -60,7 +66,9 @@

[GitHub] eolivelli commented on a change in pull request #1802: [branch 4.8] Fix compilation issue introduced by issue #1791

eolivelli commented on a change in pull request #1802: [branch 4.8] Fix compilation issue introduced by issue #1791 URL: https://github.com/apache/bookkeeper/pull/1802#discussion_r232920049 ## File path:

[GitHub] sijie commented on a change in pull request #1808: Allow to configure sticky reads

sijie commented on a change in pull request #1808: Allow to configure sticky reads URL: https://github.com/apache/bookkeeper/pull/1808#discussion_r232922098 ## File path: bookkeeper-server/src/main/java/org/apache/bookkeeper/client/PendingReadOp.java ## @@ -96,13 +96,38

[GitHub] eolivelli commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads.

eolivelli commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads. URL: https://github.com/apache/bookkeeper/pull/1804#issuecomment-438162643 Okay, we don't need this. This is an

[GitHub] eolivelli commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads.

eolivelli commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads. URL: https://github.com/apache/bookkeeper/pull/1804#issuecomment-437870131 retest this please This is an automated

[GitHub] eolivelli commented on issue #1648: [tests] update bc tests to include 4.7.1 and 4.7.2

eolivelli commented on issue #1648: [tests] update bc tests to include 4.7.1 and 4.7.2 URL: https://github.com/apache/bookkeeper/pull/1648#issuecomment-437869927 We can wait for 4.8.1 release and improve this change This is

Build failed in Jenkins: bookkeeper_release_branch_48_java9 #89

See Changes: [sijie] Fixed Auth with v2 protocol [sijie] Fix bugs in DefaultEnsemblePlacementPolicy -- [...truncated 507.23 KB...]

Build failed in Jenkins: bookkeeper_release_nightly_snapshot #192

See -- Started by timer [EnvInject] - Loading node environment variables. Building remotely on H33 (ubuntu xenial) in workspace

[GitHub] eolivelli closed pull request #1800: Issue 1799: bkctl is broken with default build options

eolivelli closed pull request #1800: Issue 1799: bkctl is broken with default build options URL: https://github.com/apache/bookkeeper/pull/1800 This is a PR merged from a forked repository. As GitHub hides the original diff on merge, it is displayed below for the sake of provenance: As

[GitHub] eolivelli closed issue #1799: bkctl is broken with default build options

eolivelli closed issue #1799: bkctl is broken with default build options URL: https://github.com/apache/bookkeeper/issues/1799 This is an automated message from the Apache Git Service. To respond to the message, please log

Build failed in Jenkins: bookkeeper_postcommit_master_python #34

See Changes: [guosijie] Fixed Auth with v2 protocol [guosijie] Fix indentation in BP-34 doc [guosijie] Fix bugs in DefaultEnsemblePlacementPolicy --

[GitHub] eolivelli commented on issue #1802: [branch 4.8] Fix compilation issue introduced by issue #1791

eolivelli commented on issue #1802: [branch 4.8] Fix compilation issue introduced by issue #1791 URL: https://github.com/apache/bookkeeper/pull/1802#issuecomment-437869095 rerun integration tests This is an automated message

[GitHub] eolivelli opened a new pull request #1806: Issue 1799: bkctl is broken with default build options

eolivelli opened a new pull request #1806: Issue 1799: bkctl is broken with default build options URL: https://github.com/apache/bookkeeper/pull/1806 This is a cherry pick of #1799 on branch-4.8. Original commit is from Sijie Guo Descriptions of the changes in this PR:

[GitHub] eolivelli closed pull request #1798: [tools] package bookkeeper tools into a separated distribution

eolivelli closed pull request #1798: [tools] package bookkeeper tools into a separated distribution URL: https://github.com/apache/bookkeeper/pull/1798 This is a PR merged from a forked repository. As GitHub hides the original diff on merge, it is displayed below for the sake of

[GitHub] eolivelli commented on issue #1755: Configure Netty allocator in bookie and client

eolivelli commented on issue #1755: Configure Netty allocator in bookie and client URL: https://github.com/apache/bookkeeper/pull/1755#issuecomment-437869721 run bookkeeper-server bookie tests run bookkeeper-server client tests run bookkeeper-server tls tests run pr validation

Build failed in Jenkins: bookkeeper_release_branch_47_integrationtests #204

See Changes: [sijie] Fixed Auth with v2 protocol [sijie] Fix bugs in DefaultEnsemblePlacementPolicy -- [...truncated 11.00 KB...] root

Build failed in Jenkins: bookkeeper_release_branch_48_integrationtests #89

See Changes: [sijie] Fixed Auth with v2 protocol [sijie] Fix bugs in DefaultEnsemblePlacementPolicy -- [...truncated 447.35 KB...]

Build failed in Jenkins: bookkeeper_release_branch_48_java8 #89

See Changes: [sijie] Fixed Auth with v2 protocol [sijie] Fix bugs in DefaultEnsemblePlacementPolicy -- [...truncated 502.43 KB...]

Jenkins build is still unstable: bookkeeper_release_branch_46 #214

See

Jenkins build became unstable: bookkeeper_release_branch_47_java9 #216

See

[GitHub] eolivelli commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads.

eolivelli commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads. URL: https://github.com/apache/bookkeeper/pull/1804#issuecomment-437918755 there are still a lot of errors This is

[GitHub] sijie commented on issue #1807: [bookie] Fix sorted ledger storage rotating entry log files too frequent

sijie commented on issue #1807: [bookie] Fix sorted ledger storage rotating entry log files too frequent URL: https://github.com/apache/bookkeeper/pull/1807#issuecomment-438006347 (not sure if that's the best fix or not. any suggestions are welcome)

[GitHub] nicmichael commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads.

nicmichael commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads. URL: https://github.com/apache/bookkeeper/pull/1804#issuecomment-437985501 @sijie Yes that was the intention. I'm looking at the test failures now. Adding "|| isRecoveryRead" to

[GitHub] sijie commented on issue #1806: Issue 1799: bkctl is broken with default build options

sijie commented on issue #1806: Issue 1799: bkctl is broken with default build options URL: https://github.com/apache/bookkeeper/pull/1806#issuecomment-437988500 #1802 has to go in first. This is an automated message from

[GitHub] sijie opened a new pull request #1807: [bookie] Fix sorted ledger storage rotating entry log files too frequent

sijie opened a new pull request #1807: [bookie] Fix sorted ledger storage rotating entry log files too frequent URL: https://github.com/apache/bookkeeper/pull/1807 Descriptions of the changes in this PR: *Motivation* A strong behavior was observed when using sorted

[GitHub] reddycharan commented on a change in pull request #1807: [bookie] Fix sorted ledger storage rotating entry log files too frequent

reddycharan commented on a change in pull request #1807: [bookie] Fix sorted ledger storage rotating entry log files too frequent URL: https://github.com/apache/bookkeeper/pull/1807#discussion_r232790690 ## File path:

[GitHub] nicmichael commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads.

nicmichael commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads. URL: https://github.com/apache/bookkeeper/pull/1804#issuecomment-438045377 During recovery I suppose they should... In the absence of a recovery on a read-only LH, they don't

[GitHub] eolivelli commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads.

eolivelli commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads. URL: https://github.com/apache/bookkeeper/pull/1804#issuecomment-438010582 maybe the errors on recovery are due to the fact that reads should strictly respect the order

[GitHub] sijie commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads.

sijie commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads. URL: https://github.com/apache/bookkeeper/pull/1804#issuecomment-438053797 yeah. I think we don't this change. we should just close it.

[GitHub] sijie closed pull request #1804: [client] read submission for readonly handle should bypass OSE threads.

sijie closed pull request #1804: [client] read submission for readonly handle should bypass OSE threads. URL: https://github.com/apache/bookkeeper/pull/1804 This is a PR merged from a forked repository. As GitHub hides the original diff on merge, it is displayed below for the sake of

[GitHub] nicmichael commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads.

nicmichael commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads. URL: https://github.com/apache/bookkeeper/pull/1804#issuecomment-438056753 I *think* what we need is something like this: if (isHandleWritable() ||

[GitHub] sijie commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads.

sijie commented on issue #1804: [client] read submission for readonly handle should bypass OSE threads. URL: https://github.com/apache/bookkeeper/pull/1804#issuecomment-438058372 @nicmichael I think it is a bit risky to take `recovery` state into account. so I would rather rely on whether