Skip to content
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.

Sign up for GitHub

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

Jump to bottom

ui: jobs page no longer automatically updates progress in the browser #68109

Closed
dhartunian opened this issue Jul 27, 2021 · 1 comment · Fixed by #88159
Closed

ui: jobs page no longer automatically updates progress in the browser #68109

dhartunian opened this issue Jul 27, 2021 · 1 comment · Fixed by #88159
Assignees
@xinhaoz
Labels
C-bug Code not up to spec/doc, specs & docs deemed correct. Solution expected to change code/behavior.

Comments

@dhartunian
Copy link
Collaborator

dhartunian commented Jul 27, 2021 •
edited by exalate-issue-sync bot
Loading

Describe the problem

When the jobs page is loaded, a long-running job will not have its status automatically updated as the page is left open

To Reproduce

I created a single-node cluster, loaded up some TPCC data, and ran a local backup to create a job that took a bit of time:

cockroach start-single-node --insecure
cockroach workload init tpcc --warehouses=10
cockroach sql --insecure
> backup database tpcc into 'bkup' IN 'nodelocal://1/backup.csv';

Navigate to http://localhost:8080/#/jobs and see that the backup job doesn't update progress.

Expected behavior
Progress bar should update as the page is left open

Additional data / screenshots
Screen Shot 2021-07-27 at 11 12 36 AM

Environment:

11:02:40 ❯ cockroach version
Build Tag:        v21.1.2
Build Time:       2021/06/07 18:13:04
Distribution:     CCL
Platform:         darwin amd64 (x86_64-apple-darwin19)
Go Version:       go1.15.11
C Compiler:       Clang 10.0.0
Build Commit ID:  d9845bd3021b4f5b816bd8bcde58b90a73f0377c
Build Type:       release

Jira issue: CRDB-8872
@dhartunian dhartunian added C-bug Code not up to spec/doc, specs & docs deemed correct. Solution expected to change code/behavior. T-cluster-ui labels Jul 27, 2021
@dhartunian
Copy link
Collaborator Author

Confirm with bulk-io team that executing job details retrieval via the crdb_internal.jobs table does not count as RUs for multi-tenant applications.

xinhaoz added a commit to xinhaoz/cockroach that referenced this issue Sep 19, 2022
@xinhaoz
ui/cluster-ui: fix jobs page polling
a3d9b0e 
Fixes cockroachdb#68109

Previously, the jobs page would not poll for new data
until a re-render was triggered. This commit updates
the jobs page polling to every 10s regardless of
whether or not the rest of the page has changed.

Release note (bug fix): jobs page refreshes page data
at an interval of 10s
@xinhaoz xinhaoz mentioned this issue Sep 19, 2022
Merged
xinhaoz added a commit to xinhaoz/cockroach that referenced this issue Sep 19, 2022
@xinhaoz
ui/cluster-ui: fix jobs page polling
6f04ffd 
Fixes cockroachdb#68109

Previously, the jobs pages would not poll for new data
until a re-render was triggered. This commit updates
the jobs page polling to every 10s regardless of
whether or not the rest of the page has changed on the
overview page, and polls every 10s on the details page
if the job is in a non-terminal state.

Release note (bug fix): jobs pages refreshes data at an
interval of 10s
craig bot pushed a commit that referenced this issue Sep 20, 2022
@ZhouXing19 @fqazi
@xinhaoz @ericharmeling @maryliag @knz @nvanbenschoten
Merge #86334 #87304 #87896 #88138 #88159 #88165 #88168 #88173 #88193
4d66931 
86334: *: add restriction to running DDL with internal executors r=ajwerner,rafiss a=ZhouXing19

The current internal executor has its lifecycle, which makes it erroneous
when being used to execute DDL statements if under an outer txn. 

In this commit, we 
1. Migrated the existing DDLs with internal executor with not-nil txn to either 
`descs.CollectionFactory.TxnWithExecutor()` or `planner.WithInternalExecutor()`.
Only internal executors inited via these 2 interfaces are bounded with txn-related
metadata, and hence are allowed to run DDLs in a transactional manner.
2. Added a restriction for running DDLs with internal executor only if it's bound with
txn-related metadata.

fixes #87281

Release justification: bug fix for the internal executor
Release note: none

87304: roachtest: refactor github issue posting and expose actual vm args for reporting r=renatolabs a=smg260

The change is split into 2 commits. 
1. More strictly addresses [81846](#81846).  Also addresses [this](https://github.com/cockroachdb/cockroach/pull/87304/files#diff-1e3ec1fc5bade2df4b8681c44959be2a390f703839f30a979da98f963769c9fcR1046) existing TODO
2. Refactors functions related to posting an issue to github into a separate source file and adds tests.

These commits can be squashed but are separate at the moment for convenience. 

Resolves: #81846

Release justification: test-only change
Release note: none

87896: ci: fix validation for mixed version schema change corpus r=fqazi a=fqazi

Informs: #86381

Previously, the command for validating the mixed
version corpus was invalid. This patch fixes the validation 
command to address this, so the mixed version corpus is 
automatically uploaded.

Release note: None

88138: server: return all completed stmt diag requests r=xinhaoz a=xinhaoz

Fixes #80104

Previously, we only return statement diagnostics requests that have not yet expired. Since we use the results of this request to populate completed statement diagnostics bundles in addition to outstanding requests, completed statement diag bundles would disappear from the UI after the request expired.

This commit ensures that `StatementDiagnosticsRequests` returns all completed stmt diag requests so that we can display the complete history of completed bundles in the UI.

Release note (bug fix): completed stmt diagnostics bundles now persist in the UI in stmt diag bundle pages

88159: ui/cluster-ui: fix jobs page polling r=xinhaoz a=xinhaoz

Fixes #68109

Previously, the jobs page would not poll for new data until a re-render was triggered. This commit updates the jobs page polling to every 10s regardless of
whether or not the rest of the page has changed.

Release note (bug fix): jobs page refreshes page data at an interval of 10s

88165: sql: version gate idx recommendations in insert-stmt-stats r=ericharmeling a=ericharmeling

This commit version gates index recommendation insert in insert-stmt-stats.

Fixes #88140.

Release note: None

88168: api: increase timeout of sql api r=maryliag a=maryliag

Previously, some request using the sql-over-http
api were hitting a timeout, with the default value of 5s. This commit increases to 300s (5min) on the calls made from SQL Activity and Insights pages.

Fixes #88094

Release note: None

88173: cli/sql: put a limit on history size r=DrewKimball a=knz

Fixes #54679.

Previously, there was no limit. Some users managed to make their history run into megabyte-size, despite de-duplication, which was causing slowness.

This patch fixes it by adding a limit of 1000 entries. Sufficiently large to not be inconvenient, but sufficiently small that it prevents the history file from growing abnormally large.

Release note (cli change): The interactive SQL shell now retains a maximum of 1000 entries. There was no limit previously.

88193: kv: hoist log.ExpensiveLogEnabled outside loop in appendRefreshSpans r=arulajmani a=nvanbenschoten

This commit hoists the call to `log.ExpensiveLogEnabled` outside of the per-span loop in `txnSpanRefresher.appendRefreshSpans`. The function is too expensive to call once per span.

Release justification: low risk change that may close perf gap.

Release note: None.

Co-authored-by: Jane Xing <[email protected]>
Co-authored-by: Miral Gadani <[email protected]>
Co-authored-by: Faizan Qazi <[email protected]>
Co-authored-by: Xin Hao Zhang <[email protected]>
Co-authored-by: Eric Harmeling <[email protected]>
Co-authored-by: Marylia Gutierrez <[email protected]>
Co-authored-by: Raphael 'kena' Poss <[email protected]>
Co-authored-by: Nathan VanBenschoten <[email protected]>
@craig craig bot closed this as completed in 7ffd2e3 Sep 20, 2022
blathers-crl bot pushed a commit that referenced this issue Sep 20, 2022
@xinhaoz
ui/cluster-ui: fix jobs page polling
cd6a94d 
Fixes #68109

Previously, the jobs pages would not poll for new data
until a re-render was triggered. This commit updates
the jobs page polling to every 10s regardless of
whether or not the rest of the page has changed on the
overview page, and polls every 10s on the details page
if the job is in a non-terminal state.

Release note (bug fix): jobs pages refreshes data at an
interval of 10s
@blathers-crl blathers-crl bot mentioned this issue Sep 20, 2022
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@xinhaoz xinhaoz

Labels
C-bug Code not up to spec/doc, specs & docs deemed correct. Solution expected to change code/behavior.
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

ui/cluster-ui: fix jobs page polling xinhaoz/cockroach
3 participants
@dhartunian @zachlite @xinhaoz