+++ This bug was initially created as a clone of Bug #1162216 +++
Description of problem:
While gfs2_edit savemeta performance was improved for RHEL7.2, it still takes a long time to dump a file systems over 1TB in size. Let's explore what else can be done to improve performance of this debugging task.
FS | gfs2-utils-3.1.7-6.el7 | gfs2-utils-3.1.8-6.el7
Size | elapsed CPU% Memory | elapsed CPU% Memory
=============================================================
10GB | 0:57.72 46% 11852k | 0:15.30 10% 11012k
100GB | 16:48.19 27% 19504k | 9:30.99 3% 11140k
1TB | 3:27:41 22% 97700k | 2:32:31 2% 12672k
Just a thought - we made the default compression level in gfs2_edit savemeta to be -z 9, which means best compression level vs speed. If that caused a significant performance overhead then we could definitely reconsider the default setting, but also Support could advise customers with larger filesystems to use lower -z values when asking for metadata, before it's even fixed. Obviously the size of the metadata file will increase with faster compression levels so that will be a factor to consider. It would be interesting to see some performance comparisons between the -z levels.
Comment 7RHEL Program Management
2020-11-01 03:02:47 UTC
After evaluating this issue, there are no plans to address it further or fix it in an upcoming release. Therefore, it is being closed. If plans change such that this issue will be fixed in an upcoming release, then the bug can be reopened.