Hi Community,
The task named "Lucene Binaries Cleanup" in Daily Maintenance job is taking lot of time to complete. Below is the statistics extracted from useraudit.log file for same.
Weekly Job Run | Daily Job Run | Time Taken by LuceneBinariesCleanup Task |
| 24/07/2024, Wed | 9.31Hrs |
| 25/07/2024, Thu | 10.16Hrs |
| 26/07/2024, Fri | 9.30Hrs |
27/07/2024, Sat - Weekly Maintenance Task | 27/07/2024, Sat | 7.09Hrs |
| 28/07/2024, Sun | 7.43Hrs |
| 29/07/2024, Mon | 14.48Hrs |
| 30/07/2024, Tue | 15:20Hrs |
| 31/07/2024, Wed | 16.20Hrs |
| 01/08/2024, Thu | 05:03Hrs |
| 02/08/2024, Fri | 05:41Hrs |
03/08/2024, Sat - Weekly Maintenance Task | 03/08/2024, Sat | 05:46Hrs |
| 04/08/2024, Sun | 08.30Hrs |
| 05/08/2024, Mon | 09.34Hrs |
| 06/08/2024, Tue | 7.40Hrs |
| 07/08/2024, Wed | 9.20Hrs |
| 08/08/2024, Thu | 6.47Hrs |
| 09/08/2024, Fri | 1.50hrs |
10/08/2024, Sat - Weekly Maintenance Task | 10/08/2024, Sat | 1.47hrs |
| 11/08/2024, Sun | 2.24hrs |
| 12/08/2024, Mon | 8.56Hrs |
Below is my case,
* AEM on-premise
* Happens only on PROD Author environment, but works normally on lower environment
* Also reproducible on Cloned PROD environment, just to check if traffic is the issue.
* VersionPurge task is running weekly, but Health reports says it has to be executed daily. However, tried scheduling it on daily basis on Cloned PROD env, but in vain.
Most importantly, this job generates the .tmp files in the underlying /tmp folder and causes overflow issues
Question :
1. Is this normal for the given task to take hours?
2. Why sometime it finishes in 1-2hrs and sometime it lasts for 10-15+ hours?
3. Last but not the least, any ideas how I can optimize it?
4. Even after successful Lucene Binaries cleanup, why the next job run takes huge amount of time?
5. Gone through documentation, but didn't help much.
Because, this job consumes /tmp folder's 100% space, other tasks like package creation using /crx/packmgr OR content authoring or content publishing gets blocked.
Help suggestions are welcomed.
KR,
Prasanna
Views
Replies
Total Likes
@Tad_Reeves @aanchal-sikka @Raja_Reddy @BrianKasingli Could you kindly take a look at this question and provide your thoughts? Your insights would be greatly appreciated.
Hi @kautuk_sahni Thanks for looking into it. I got a fix from Adobe Support(Rohith Venati) on this..Default value of the property "blobTrackSnapshotIntervalInSecs" from "org.apache.jackrabbit.oak.segment.SegmentNodeStoreService.config" was the root cause. I was advised to set it to ZERO, tested this on multiple times on multiple Clones PROD AUthor environments and moved to PROD last week.
Views
Replies
Total Likes
Views
Likes
Replies