[MCOL-1453] mcsapi doesn't detect when WriteEngineServer has gone away Created: 2018-06-05 Updated: 2018-07-02 Resolved: 2018-06-27 |
|
| Status: | Closed |
| Project: | MariaDB ColumnStore |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | 1.1.6 |
| Type: | Bug | Priority: | Major |
| Reporter: | Andrew Hutchings (Inactive) | Assignee: | Daniel Lee (Inactive) |
| Resolution: | Fixed | Votes: | 0 |
| Labels: | None | ||
| Sprint: | 2018-12, 2018-13 |
| Description |
|
I suspect we aren't catching the libuv event for this. But killing WriteEngineServer seems to send the API into a loop. |
| Comments |
| Comment by Andrew Hutchings (Inactive) [ 2018-06-14 ] |
|
Please hold the review until after 1.1.5 has been released. For QA: Best way to trigger this is to execute test/mcol1408 and after about 20 seconds run "killall WriteEngineServ". Before the patch the execution will get stuck in an endless loop, using 100% CPU time for a core. After the patch an error will occur and the test will fail and exit. |
| Comment by Daniel Lee (Inactive) [ 2018-06-27 ] |
|
Build verified: GitHub source /root/columnstore/mariadb-columnstore-server Merge pull request #113 from mariadb-corporation/davidhilldallas-patch-3 update readme /root/columnstore/mariadb-columnstore-server/mariadb-columnstore-engine Merge pull request #508 from mariadb-corporation/ /root/mariadb-columnstore-tools update version diff --git a/VERSION b/VERSION Merge pull request #81 from mariadb-corporation/ Verified per instruction for QA. |