[kudu-CR] Do not enter ALTERING state just for replication factor changed.
Hello Alexey Serbin, Kudu Jenkins, I'd like you to reexamine a change. Please visit http://gerrit.cloudera.org:8080/20830 to look at the new patch set (#4). Change subject: Do not enter ALTERING state just for replication factor changed. .. Do not enter ALTERING state just for replication factor changed. If we change the replication factor of a table without any tablets, the table will stay in ALTERING state since exiting this state needs tablet reports from the tablets of this table. for now the condition of entering ALTERING state is as below: has_metadata_changes && (table->num_tablets() > tablets_to_drop.size() || num_replicas_changed); Actually even if the number of replicas has been changed, the former condition is also needed. So rollback the code to: has_metadata_changes && table->num_tablets() > tablets_to_drop.size(); Change-Id: I1d20da7c0dd5912790aaa46e9fff366b2973d7a4 --- M src/kudu/integration-tests/alter_table-test.cc M src/kudu/master/catalog_manager.cc 2 files changed, 39 insertions(+), 2 deletions(-) git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/30/20830/4 -- To view, visit http://gerrit.cloudera.org:8080/20830 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: newpatchset Gerrit-Change-Id: I1d20da7c0dd5912790aaa46e9fff366b2973d7a4 Gerrit-Change-Number: 20830 Gerrit-PatchSet: 4 Gerrit-Owner: Song Jiacheng Gerrit-Reviewer: Alexey Serbin Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Reviewer: Song Jiacheng
[kudu-CR] [compaction/test] Add tests to generate heavy rowset compaction
Hello Tidy Bot, Kudu Jenkins, I'd like you to reexamine a change. Please visit http://gerrit.cloudera.org:8080/20816 to look at the new patch set (#6). Change subject: [compaction/test] Add tests to generate heavy rowset compaction .. [compaction/test] Add tests to generate heavy rowset compaction This change implements a better approach to test out high memory compaction case. There is an existing patch: https://gerrit.cloudera.org/#/c/19278/ that uses different approach which does not scale well because of excessive memory usage by client. Patch contains two tests: 1. TestRowSetCompactionProceedWithNoBudgetingConstraints (generates deltas with memory requirements under budget) 2. TestRowSetCompactionSkipWithBudgetingConstraints (generates deltas with memory requirements over budget) and a helper function: GenHighMemConsumptionDeltas Using helper function, callers can generate deltas of different sizes as per test needs. The size_factor can be used to achieve that. Increasing size_factor by 1 adds 10MB worth of deltas as far as rowset compaction memory consumption is concerned. Change-Id: I1996558e71c49314c6acf12faf854c796548318c --- M src/kudu/tablet/compaction-test.cc 1 file changed, 142 insertions(+), 0 deletions(-) git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/16/20816/6 -- To view, visit http://gerrit.cloudera.org:8080/20816 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: newpatchset Gerrit-Change-Id: I1996558e71c49314c6acf12faf854c796548318c Gerrit-Change-Number: 20816 Gerrit-PatchSet: 6 Gerrit-Owner: Ashwani Raina Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Reviewer: Tidy Bot (241)
[kudu-CR] [compaction] Add memory estimation unit test
Ashwani Raina has posted comments on this change. ( http://gerrit.cloudera.org:8080/20787 ) Change subject: [compaction] Add memory estimation unit test .. Patch Set 6: (1 comment) http://gerrit.cloudera.org:8080/#/c/20787/6//COMMIT_MSG Commit Message: http://gerrit.cloudera.org:8080/#/c/20787/6//COMMIT_MSG@9 PS6, Line 9: FLAGS_rowset_compaction_memory_estimate_enabled was always set to : false during testing, so this part of the code was not executed in : testing environment > Indeed! I recall that when I was working on 1556a353e, I verified the code Hi Alexey, Adam, Just a side note - while the test added here does go through the memory budgeting estimation code added in 1556a353e, it relies on MockDiskRowSet and its size that is set by the test to ensure it is more than the budget. That does the trick and test out the budgeting checks with mock rowsets. Additionally, I have added unit tests to test out the same budgeting checks with actual workload that generates deltas with specified memory requirements for compaction. You can check it out here: https://gerrit.cloudera.org/#/c/20816/ -- To view, visit http://gerrit.cloudera.org:8080/20787 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I5397c7e7bf29942a610b0b8203b6918811c49bbb Gerrit-Change-Number: 20787 Gerrit-PatchSet: 6 Gerrit-Owner: Ádám Bakai Gerrit-Reviewer: Abhishek Chennaka Gerrit-Reviewer: Alexey Serbin Gerrit-Reviewer: Ashwani Raina Gerrit-Reviewer: Attila Bukor Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Reviewer: Ádám Bakai Gerrit-Comment-Date: Wed, 03 Jan 2024 06:23:50 + Gerrit-HasComments: Yes
[kudu-CR](gh-pages) Add blog post about Prometheus/Grafana
Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/20814 ) Change subject: Add blog post about Prometheus/Grafana .. Patch Set 1: (3 comments) Thank you for working on a blog post! That's been quite long time since last blog posted at Apache Kudu website. http://gerrit.cloudera.org:8080/#/c/20814/1/_posts/2023-12-18-kudu-prometheus-grafana.md File _posts/2023-12-18-kudu-prometheus-grafana.md: http://gerrit.cloudera.org:8080/#/c/20814/1/_posts/2023-12-18-kudu-prometheus-grafana.md@18 PS1, Line 18: JMX format I'm not sure what 'JMX format' means, but I might be missing something. Could you provide a few references, so I could learn about the 'JMX format'? As we discussed offline already, I don't think mentioning JMX framework is quite relevant in this context. If there is an idea to point that there is no need to perform any transformations for Kudu metrics produced by the '/metrics' kudu-master/kudu-tserver endpoints, let's do so in a more generic way, maybe? If it's desirable to add more specifics (which I'm not sure about, frankly: why to so that in this blog?), maybe mention a few common ways how people usually build bridges between metric sources of arbitrary formats and Prometheus. But mentioning only the JMX framework (which is Java-specific) seems to be irrelevant both to Prometheus as a consumer (Go) and Kudu server-side as a producer (C++). http://gerrit.cloudera.org:8080/#/c/20814/1/_posts/2023-12-18-kudu-prometheus-grafana.md@17 PS1, Line 17: to add : custom jars Add custom JARs where? http://gerrit.cloudera.org:8080/#/c/20814/1/_posts/2023-12-18-kudu-prometheus-grafana.md@172 PS1, Line 172: can create a script which parses the TServer addresses Consider mentioning KUDU-3538: ideally, Kudu masters could implement HTTP Service Discovery for Prometheus. Once implemented, it will be enough to specify the SD URL(s) in the Prometheus config file. Alternatively, you could try to implement that trivial enhancement first, and then update this blog post :) -- To view, visit http://gerrit.cloudera.org:8080/20814 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: gh-pages Gerrit-MessageType: comment Gerrit-Change-Id: I85878378f883b2cc1f8136bde2c9f0db7ceccd8f Gerrit-Change-Number: 20814 Gerrit-PatchSet: 1 Gerrit-Owner: Marton Greber Gerrit-Reviewer: Alexey Serbin Gerrit-Reviewer: Attila Bukor Gerrit-Reviewer: Marton Greber Gerrit-Reviewer: Zoltan Chovan Gerrit-Comment-Date: Wed, 03 Jan 2024 04:10:48 + Gerrit-HasComments: Yes
[kudu-CR](gh-pages) Add blog post about Prometheus/Grafana
Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/20814 ) Change subject: Add blog post about Prometheus/Grafana .. Patch Set 1: (2 comments) http://gerrit.cloudera.org:8080/#/c/20814/1/_posts/2023-12-18-kudu-prometheus-grafana.md File _posts/2023-12-18-kudu-prometheus-grafana.md: http://gerrit.cloudera.org:8080/#/c/20814/1/_posts/2023-12-18-kudu-prometheus-grafana.md@24 PS1, Line 24: Do you think it's worth mentioning that currently Kudu exports only server-level metrics in Prometheus format? I'm concerned that people might read this blog and think that following these steps they are about to get all the Kudu metrics (as available at /metrics end-point), but just in Prometheus format. However, that's not so -- tablet-level metrics aren't available in Prometheus format as of now. To avoid disappointment and confusion, I'd suggest to explicitly mention this in this blog, adding that future Kudu versions should eventually expose all the metrics available. http://gerrit.cloudera.org:8080/#/c/20814/1/_posts/2023-12-18-kudu-prometheus-grafana.md@33 PS1, Line 33: ## Prometheus Configuration for Kudu As a side note, it would be much easier to integrate with Prometheus if Kudu provided HTTP Service Discovery mechanismter. I filed KUDU-3538 to track that. -- To view, visit http://gerrit.cloudera.org:8080/20814 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: gh-pages Gerrit-MessageType: comment Gerrit-Change-Id: I85878378f883b2cc1f8136bde2c9f0db7ceccd8f Gerrit-Change-Number: 20814 Gerrit-PatchSet: 1 Gerrit-Owner: Marton Greber Gerrit-Reviewer: Alexey Serbin Gerrit-Reviewer: Attila Bukor Gerrit-Reviewer: Marton Greber Gerrit-Reviewer: Zoltan Chovan Gerrit-Comment-Date: Wed, 03 Jan 2024 01:26:23 + Gerrit-HasComments: Yes
[kudu-CR] [compaction] Add memory estimation unit test
Alexey Serbin has posted comments on this change. (
http://gerrit.cloudera.org:8080/20787 )
Change subject: [compaction] Add memory estimation unit test
..
Patch Set 6:
(10 comments)
http://gerrit.cloudera.org:8080/#/c/20787/6//COMMIT_MSG
Commit Message:
http://gerrit.cloudera.org:8080/#/c/20787/6//COMMIT_MSG@9
PS6, Line 9: FLAGS_rowset_compaction_memory_estimate_enabled was always set to
: false during testing, so this part of the code was not executed in
: testing environment
Indeed! I recall that when I was working on 1556a353e, I verified the code
worked with the data captured at a real system, but then I cleaned up the units
for the newly introduced flags, and screwed up, forgetting to update the factor
as needed. Ashwani fixed the issue with ae7b08c00, but it's great to
eventually have a test for this.
Thank you very much for adding the test!
http://gerrit.cloudera.org:8080/#/c/20787/6/src/kudu/tablet/compaction_policy-test.cc
File src/kudu/tablet/compaction_policy-test.cc:
http://gerrit.cloudera.org:8080/#/c/20787/6/src/kudu/tablet/compaction_policy-test.cc@648
PS6, Line 648: TestMemoryEstimation
nit: 'Test' is already a part of the name of the test itself
('TestCompactionPolicy'); consider not duplicating it in the name of the
scenario, i.e. renaming this into 'MemoryEstimation'
http://gerrit.cloudera.org:8080/#/c/20787/6/src/kudu/tablet/compaction_policy-test.cc@651
PS6, Line 651: 2
Why is this magic '2' factor? Is it possible to get rid of it, correspondingly
updating the sizes of mock rowsets below?
http://gerrit.cloudera.org:8080/#/c/20787/6/src/kudu/tablet/compaction_policy-test.cc@651
PS6, Line 651: *1024*1024
style nit for here and below: separate operation symbol from the operands by
space
http://gerrit.cloudera.org:8080/#/c/20787/6/src/kudu/tablet/compaction_policy-test.cc@658
PS6, Line 658: 20
Is it crucial to have this factor 20? Isn't the compaction size estimation
good enough to work as expected with a couple of rowsets each of
memory_limit_hard_bytes size?
http://gerrit.cloudera.org:8080/#/c/20787/6/src/kudu/tablet/compaction_policy-test.cc@663
PS6, Line 663: picked.size(), 2
nit for here and below: the expected size comes first, otherwise it's harder to
decipher the error message when the assertion triggers
http://gerrit.cloudera.org:8080/#/c/20787/6/src/kudu/tablet/compaction_policy-test.cc@672
PS6, Line 672: for (const auto& str : post_capture_logs.logged_msgs()) {
: sum += str;
: }
: ASSERT_STR_CONTAINS(sum, "removed from compaction input due
to memory constraints");
Consider using JoinStrings() from strings/join.h
Isn't it enough just to search for the pattern in each logged message? I'd
think there isn't any concurrency involved here, so it would work, no?
http://gerrit.cloudera.org:8080/#/c/20787/6/src/kudu/tablet/diskrowset.cc
File src/kudu/tablet/diskrowset.cc:
http://gerrit.cloudera.org:8080/#/c/20787/6/src/kudu/tablet/diskrowset.cc@794
PS6, Line 794: uint64_t DiskRowSet::OnDiskDeltaSize() const {
Once this new method is introduced, why not to update the code in
Tablet::DoMergeCompactionOrFlush(), removing the down-casting?
http://gerrit.cloudera.org:8080/#/c/20787/6/src/kudu/tablet/rowset.cc
File src/kudu/tablet/rowset.cc:
http://gerrit.cloudera.org:8080/#/c/20787/6/src/kudu/tablet/rowset.cc@266
PS6, Line 266: const shared_ptr
style nit: the ampersand sticks to the type, not the variable
Also, consider using 'const auto&' instead.
http://gerrit.cloudera.org:8080/#/c/20787/6/src/kudu/tablet/rowset.cc@266
PS6, Line 266: new_rowsets_
What about 'old_rowsets_'? Those also consume some disk space, no?
--
To view, visit http://gerrit.cloudera.org:8080/20787
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I5397c7e7bf29942a610b0b8203b6918811c49bbb
Gerrit-Change-Number: 20787
Gerrit-PatchSet: 6
Gerrit-Owner: Ádám Bakai
Gerrit-Reviewer: Abhishek Chennaka
Gerrit-Reviewer: Alexey Serbin
Gerrit-Reviewer: Ashwani Raina
Gerrit-Reviewer: Attila Bukor
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Ádám Bakai
Gerrit-Comment-Date: Tue, 02 Jan 2024 21:23:08 +
Gerrit-HasComments: Yes
[kudu-CR] Do not enter ALTERING state just for replication factor changed.
Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/20830 ) Change subject: Do not enter ALTERING state just for replication factor changed. .. Patch Set 3: Code-Review+1 (5 comments) > (2 comments) > > > Patch Set 2: > > > > (3 comments) > > Thanks for the review. > I have added a test that fails without the fix. Actually without > the fix, it will time out while altering since the state of table > is altering forever. Thank you for adding a test for that! Overall the change looks to me, just a few nits. http://gerrit.cloudera.org:8080/#/c/20830/2//COMMIT_MSG Commit Message: http://gerrit.cloudera.org:8080/#/c/20830/2//COMMIT_MSG@17 PS2, Line 17: d, the former : condition is also needed. So > Sorry, I am not sure. The words here are trying to say that we can not make Sure: you are free to keep the wording as is, whatever serves your purpose best. I just thought that the point was to explain why this change wasn't breaking the functionality of changing the replication factor for tables that have tablets. But it seems you were trying to emphasize some other point :) http://gerrit.cloudera.org:8080/#/c/20830/3/src/kudu/integration-tests/alter_table-test.cc File src/kudu/integration-tests/alter_table-test.cc: http://gerrit.cloudera.org:8080/#/c/20830/3/src/kudu/integration-tests/alter_table-test.cc@2652 PS3, Line 2652: unique_ptr table_alterer; : table_alterer.reset(client_->NewTableAlterer(kTableName)); nit: there is no need to reset the instance, it could be done in one-liner unique_ptr table_alterer(client_->NewTableAlterer(kTableName)); http://gerrit.cloudera.org:8080/#/c/20830/3/src/kudu/integration-tests/alter_table-test.cc@2656 PS3, Line 2656: Does it make sense to add an extra check, making sure there are any tables in the table before trying to update its replication factor? It's possible to do so using the itest::GetTableLocations() utility function (you could check delete_tablet-itest.cc as a reference for itest::GetTableLocations() usage). http://gerrit.cloudera.org:8080/#/c/20830/3/src/kudu/integration-tests/alter_table-test.cc@2661 PS3, Line 2661: CHECK_OK Why not ASSERT_OK()? http://gerrit.cloudera.org:8080/#/c/20830/3/src/kudu/integration-tests/alter_table-test.cc@2667 PS3, Line 2667: CHECK_OK ditto: use ASSERT_OK instead of CHECK_OK? -- To view, visit http://gerrit.cloudera.org:8080/20830 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I1d20da7c0dd5912790aaa46e9fff366b2973d7a4 Gerrit-Change-Number: 20830 Gerrit-PatchSet: 3 Gerrit-Owner: Song Jiacheng Gerrit-Reviewer: Alexey Serbin Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Reviewer: Song Jiacheng Gerrit-Comment-Date: Tue, 02 Jan 2024 20:29:39 + Gerrit-HasComments: Yes
[kudu-CR] [compaction/test] Add tests to generate heavy rowset compaction
Hello Tidy Bot, Kudu Jenkins, I'd like you to reexamine a change. Please visit http://gerrit.cloudera.org:8080/20816 to look at the new patch set (#5). Change subject: [compaction/test] Add tests to generate heavy rowset compaction .. [compaction/test] Add tests to generate heavy rowset compaction This change implements a better approach to test out high memory compaction case. There is an existing patch: https://gerrit.cloudera.org/#/c/19278/ that uses different approach which does not scale well because of excessive memory usage by client. Patch contains two tests: 1. TestRowSetCompactionProceedWithNoBudgetingConstraints (generates deltas with memory requirements under budget) 2. TestRowSetCompactionSkipWithBudgetingConstraints (generates deltas with memory requirements over budget) and a helper function: GenHighMemConsumptionDeltas Using helper function, callers can generate deltas of different sizes as per test needs. The size_factor can be used to achieve that. Increasing size_factor by 1 adds 10MB worth of deltas as far as rowset compaction memory consumption is concerned. Change-Id: I1996558e71c49314c6acf12faf854c796548318c --- M src/kudu/tablet/compaction-test.cc 1 file changed, 142 insertions(+), 0 deletions(-) git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/16/20816/5 -- To view, visit http://gerrit.cloudera.org:8080/20816 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: newpatchset Gerrit-Change-Id: I1996558e71c49314c6acf12faf854c796548318c Gerrit-Change-Number: 20816 Gerrit-PatchSet: 5 Gerrit-Owner: Ashwani Raina Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Reviewer: Tidy Bot (241)
[kudu-CR] [compaction/test] Add tests to generate heavy rowset compaction
Hello Tidy Bot, Kudu Jenkins, I'd like you to reexamine a change. Please visit http://gerrit.cloudera.org:8080/20816 to look at the new patch set (#4). Change subject: [compaction/test] Add tests to generate heavy rowset compaction .. [compaction/test] Add tests to generate heavy rowset compaction This change implements a better approach to test out high memory compaction case. There is an existing patch: https://gerrit.cloudera.org/#/c/19278/ that uses different approach which does not scale well because of excessive memory usage by client. Patch contains two tests: 1. TestRowSetCompactionProceedWithNoBudgetingConstraints (generates deltas with memory requirements under budget) 2. TestRowSetCompactionSkipWithBudgetingConstraints (generates deltas with memory requirements over budget) and a helper function: GenHighMemConsumptionDeltas Using helper function, callers can generate deltas of different sizes as per test needs. The size_factor can be used to achieve that. Increasing size_factor by 1 adds 10MB worth of deltas as far as rowset compaction memory consumption is concerned. Change-Id: I1996558e71c49314c6acf12faf854c796548318c --- M src/kudu/tablet/compaction-test.cc 1 file changed, 169 insertions(+), 0 deletions(-) git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/16/20816/4 -- To view, visit http://gerrit.cloudera.org:8080/20816 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: newpatchset Gerrit-Change-Id: I1996558e71c49314c6acf12faf854c796548318c Gerrit-Change-Number: 20816 Gerrit-PatchSet: 4 Gerrit-Owner: Ashwani Raina Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Reviewer: Tidy Bot (241)
[kudu-CR] [compaction/test] Add tests to generate heavy rowset compaction
Hello Kudu Jenkins, I'd like you to reexamine a change. Please visit http://gerrit.cloudera.org:8080/20816 to look at the new patch set (#3). Change subject: [compaction/test] Add tests to generate heavy rowset compaction .. [compaction/test] Add tests to generate heavy rowset compaction This change implements a better approach to test out high memory compaction case. There is an existing patch: https://gerrit.cloudera.org/#/c/19278/ that uses different approach which does not scale well because of excessive memory usage by client. Patch contains two tests: 1. TestRowSetCompactionProceedWithNoBudgetingConstraints (generates deltas with memory requirements under budget) 2. TestRowSetCompactionSkipWithBudgetingConstraints (generates deltas with memory requirements over budget) and a helper function: GenHighMemConsumptionDeltas Using helper function, callers can generate deltas of different sizes as per test needs. The size_factor can be used to achieve that. Increasing size_factor by 1 adds 10MB worth of deltas as far as rowset compaction memory consumption is concerned. Change-Id: I1996558e71c49314c6acf12faf854c796548318c --- M src/kudu/tablet/compaction-test.cc 1 file changed, 169 insertions(+), 0 deletions(-) git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/16/20816/3 -- To view, visit http://gerrit.cloudera.org:8080/20816 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: newpatchset Gerrit-Change-Id: I1996558e71c49314c6acf12faf854c796548318c Gerrit-Change-Number: 20816 Gerrit-PatchSet: 3 Gerrit-Owner: Ashwani Raina Gerrit-Reviewer: Kudu Jenkins (120)
[kudu-CR] [fs] Simplify the data directory inheritance hierarchy
Ádám Bakai has posted comments on this change. ( http://gerrit.cloudera.org:8080/20833 ) Change subject: [fs] Simplify the data directory inheritance hierarchy .. Patch Set 3: Code-Review+1 -- To view, visit http://gerrit.cloudera.org:8080/20833 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I2d7059e32c7fd8444256ea031bebd153b318a5a3 Gerrit-Change-Number: 20833 Gerrit-PatchSet: 3 Gerrit-Owner: Yingchun Lai Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Reviewer: Wang Xixu <1450306...@qq.com> Gerrit-Reviewer: Yifan Zhang Gerrit-Reviewer: Yingchun Lai Gerrit-Reviewer: Yuqi Du Gerrit-Reviewer: Ádám Bakai Gerrit-Comment-Date: Tue, 02 Jan 2024 11:29:58 + Gerrit-HasComments: No
[kudu-CR](gh-pages) Add blog post about Prometheus/Grafana
Attila Bukor has posted comments on this change. ( http://gerrit.cloudera.org:8080/20814 ) Change subject: Add blog post about Prometheus/Grafana .. Patch Set 1: (4 comments) http://gerrit.cloudera.org:8080/#/c/20814/1/_posts/2023-12-18-kudu-prometheus-grafana.md File _posts/2023-12-18-kudu-prometheus-grafana.md: http://gerrit.cloudera.org:8080/#/c/20814/1/_posts/2023-12-18-kudu-prometheus-grafana.md@44 PS1, Line 44: # Set the interval at which Prometheus will scrape metrics for this job (5 seconds in this case) nit: break the line http://gerrit.cloudera.org:8080/#/c/20814/1/_posts/2023-12-18-kudu-prometheus-grafana.md@55 PS1, Line 55: - targets: ['localhost:8765', 'localhost:8767', 'localhost:8769'] how about using master1.example.com:8765 etc or some such instead of varying the ports? I think it's best to use examples similar to real-word setups rather than localhost/dev setups. http://gerrit.cloudera.org:8080/#/c/20814/1/_posts/2023-12-18-kudu-prometheus-grafana.md@63 PS1, Line 63: - targets: ['localhost:9871', 'localhost:9873', 'localhost:9875'] same here http://gerrit.cloudera.org:8080/#/c/20814/1/_posts/2023-12-18-kudu-prometheus-grafana.md@93 PS1, Line 93: do we need the div etc here? can we get away with markdown tags only? -- To view, visit http://gerrit.cloudera.org:8080/20814 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: gh-pages Gerrit-MessageType: comment Gerrit-Change-Id: I85878378f883b2cc1f8136bde2c9f0db7ceccd8f Gerrit-Change-Number: 20814 Gerrit-PatchSet: 1 Gerrit-Owner: Marton Greber Gerrit-Reviewer: Alexey Serbin Gerrit-Reviewer: Attila Bukor Gerrit-Reviewer: Marton Greber Gerrit-Reviewer: Zoltan Chovan Gerrit-Comment-Date: Tue, 02 Jan 2024 10:07:14 + Gerrit-HasComments: Yes
[kudu-CR] [fs] Simplify the data directory inheritance hierarchy
Yifan Zhang has posted comments on this change. ( http://gerrit.cloudera.org:8080/20833 ) Change subject: [fs] Simplify the data directory inheritance hierarchy .. Patch Set 3: Code-Review+1 -- To view, visit http://gerrit.cloudera.org:8080/20833 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I2d7059e32c7fd8444256ea031bebd153b318a5a3 Gerrit-Change-Number: 20833 Gerrit-PatchSet: 3 Gerrit-Owner: Yingchun Lai Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Reviewer: Wang Xixu <1450306...@qq.com> Gerrit-Reviewer: Yifan Zhang Gerrit-Reviewer: Yingchun Lai Gerrit-Reviewer: Yuqi Du Gerrit-Comment-Date: Tue, 02 Jan 2024 09:18:20 + Gerrit-HasComments: No
