-
Type: Bug
-
Resolution: Gone away
-
Priority: Major - P3
-
None
-
Affects Version/s: None
-
Component/s: None
-
None
-
Environment:Centos7, WT 10.0.0
-
3
-
Storage - Ra 2021-11-29, Storage - Ra 2021-12-13
Hi there,
We are doing some testing with the new wiredtiger 10.0.0 and we have encountered some issues with reopening the connection to the wiredtiger DB.
[1637046703:105009][2351769:0x7f8277ad73c0], wt, file:r6CJAHkcscZQvemdL8RLMThIxoU3CyLEGs5DAQRRo1s_full.wt, txn rollback_to_stable: __posix_open_file, 808: /var/lib/wiredtiger/r6CJAHkcscZQvemdL8RLMThIxoU3CyLEGs5DAQRRo1s_full.wt: handle-open: open: Too many open files
[1637046703:105204][2351769:0x7f8277ad73c0], wt, txn-recover: __wt_txn_recover, 1054: Recovery failed: Too many open files
[1637046703:118307][2351769:0x7f8277ad73c0], wt, connection: __wt_cache_destroy, 364: cache server: exiting with 1103 pages in memory and 0 pages evicted
[1637046703:118342][2351769:0x7f8277ad73c0], wt, connection: __wt_cache_destroy, 367: cache server: exiting with 1937629 image bytes in memory
[1637046703:118347][2351769:0x7f8277ad73c0], wt, connection: __wt_cache_destroy, 370: cache server: exiting with 2122738 bytes in memory
[1637046703:118351][2351769:0x7f8277ad73c0], wt, connection: __wt_cache_destroy, 376: cache server: exiting with 352 bytes dirty and 1 pages dirty
wt: WT_PANIC: WiredTiger library panic
Setting the service 'LimitNOFILE' and 'LimitNPROC' to high value solves the issue.
But we have noticed that all the files that are in the database directory are being opened when we are restarting our service (and opening the WT connection).
Is there any reason for the use of so many open fds when opening the connection?
We didn't encounter this issue on version 3.2.1.
Thank you for your help.
- related to
-
WT-7442 RTS to open dhandle only when the dhandle has unstable updates
- Closed