Yingchun Lai has posted comments on this change. ( http://gerrit.cloudera.org:8080/17942 )
Change subject: [LBM] Speed up server bootstrap by using multi-thread to compact containers ...................................................................... Patch Set 2: (3 comments) http://gerrit.cloudera.org:8080/#/c/17942/2/src/kudu/fs/log_block_manager.cc File src/kudu/fs/log_block_manager.cc: http://gerrit.cloudera.org:8080/#/c/17942/2/src/kudu/fs/log_block_manager.cc@2106 PS2, Line 2106: .set_max_threads(dd_manager_->dirs().size()) > how large can dd_manager_->dirs().size() get in production and shouldn't we The largest number of data dirs I've ever see in production is 24, I think it's rare to exceed 100 in real world production, that's OK to create such a thread pool. http://gerrit.cloudera.org:8080/#/c/17942/2/src/kudu/fs/log_block_manager.cc@2977 PS2, Line 2977: first_fatal_error = s.CloneAndPrepend( > Is first_fatal_error always guaranteed to contain the first encountered err Yes, it's always the first encountered error. I used a std::atomic<bool>'s compare_exchange_strong to guarante it. http://gerrit.cloudera.org:8080/#/c/17942/2/src/kudu/fs/log_block_manager.cc@2992 PS2, Line 2992: LOG_AND_RETURN(WARNING, first_fatal_error); > perhaps, it would be beneficial to aggregate errors and log all errors, if Because short circuit returned once error happend, so we can't collect all of the perhaps errors. -- To view, visit http://gerrit.cloudera.org:8080/17942 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: Ie48211d9e8c1d74e520fcb04df25c1d681261bb5 Gerrit-Change-Number: 17942 Gerrit-PatchSet: 2 Gerrit-Owner: Yingchun Lai <[email protected]> Gerrit-Reviewer: Abhishek Chennaka <[email protected]> Gerrit-Reviewer: Greg Solovyev <[email protected]> Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Reviewer: Yingchun Lai <[email protected]> Gerrit-Comment-Date: Wed, 20 Oct 2021 02:35:37 +0000 Gerrit-HasComments: Yes
