[MCOL-1828] Failed cpimport, due to a Connection Timed Out, leads to one having to restartSystem Created: 2018-10-23 Updated: 2023-07-02 Resolved: 2023-07-02 |
|
| Status: | Closed |
| Project: | MariaDB ColumnStore |
| Component/s: | N/A |
| Affects Version/s: | 1.1.6 |
| Fix Version/s: | Icebox |
| Type: | Bug | Priority: | Critical |
| Reporter: | Chris Calender (Inactive) | Assignee: | Roman |
| Resolution: | Won't Do | Votes: | 1 |
| Labels: | None | ||
| Issue Links: |
|
||||||||
| Description |
|
A failed cpimort left the system in a state where a full restart is needed: /usr/local/mariadb/columnstore/bin/cpimport db table /path/to/file.tab -s '\t' -m 1 This failed cpimport (which is apparently due to some timeout between UM and PM3) left some table locks that could not be removed, and the entire system needed restarted. So this is a request to see if we can code something to help recover from that. |
| Comments |
| Comment by Roman [ 2018-12-07 ] |
|
Quick research on this issue gave me an impression that cpimport is stuck in an infinite loop in WESDHandler::checkForRespMsgs() from writeengine/splitter/we_sdhandler.cpp. There must be at least two different causes for that to happen b/c locks mentioned in support #23268 were taken both in CLEANUP and LOADING phases. |
| Comment by Todd Stoffel (Inactive) [ 2023-07-02 ] |
|
The "create date" on this ticket is pre-convergence with MariaDB server. If the issue still exists in a modern version of the engine/plugin please submit a new ticket. |