[SERVER-64660] Report in serverStatus number of file descriptors used by the query system outside of the storage engine Created: 18/Mar/22 Updated: 30/Jan/24 |
|
| Status: | Open |
| Project: | Core Server |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Task | Priority: | Major - P3 |
| Reporter: | Eric Sedor | Assignee: | Backlog - Query Execution |
| Resolution: | Unresolved | Votes: | 0 |
| Labels: | query-offsite | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||
| Assigned Teams: |
Query Execution
|
||||||||
| Sprint: | QE 2023-02-20, QE 2023-03-06, QE 2023-03-20, QE 2023-04-03, QE 2023-04-17 | ||||||||
| Participants: | |||||||||
| Description |
|
Currently, nodes can reach a "too many open files" state with a low number of connections and low wiredTiger data handle usage. Reporting file descriptors used by the aggregation framework will help close a diagnostic gap in file descriptor use. Ideally, record this as two counters:
|
| Comments |
| Comment by Eric Sedor [ 08/Apr/22 ] |
|
Since a large number of aggregations could occur concurrently, I think counters are warranted here; added to the description. This is about any use of file descriptors by the aggregation framework. I'm not certain all that use is (or will be) through the external sorter, so I'm trying not to be over-specific. |
| Comment by Bruce Lucas (Inactive) [ 23/Mar/22 ] |
|
What is the specific request - to report the current number of file descriptors currently in use across all aggregations currently running at the point where serverStatus is called? This would be useful in FTDC if the offending aggregations take substantially more than 1 second, which I think is likely to be the case, especially for an aggregation that uses a large number of files. Also, is this really about aggregations, or is it about file usage by anything that uses the external sorter? |