Bug 2224234 - client: do not send metrics until the MDS rank is ready
Summary: client: do not send metrics until the MDS rank is ready
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat Ceph Storage
Classification: Red Hat Storage
Component: CephFS
Version: 5.3
Hardware: Unspecified
OS: Unspecified
unspecified
medium
Target Milestone: ---
: 5.3z5
Assignee: Xiubo Li
QA Contact: Hemanth Kumar
Akash Raj
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2023-07-20 08:41 UTC by Xiubo Li
Modified: 2023-08-28 09:41 UTC (History)
7 users (show)

Fixed In Version: ceph-16.2.10-198.el8cp
Doc Type: Bug Fix
Doc Text:
.MDSs no longer drop the client requests and slow down Previously, once the clients were successfully reconnected they would try to send the metrics to MDSs for handling even before the MDSs were not ready. Due to this, whenever there were a lot of clients and a lot of known requests had to be replayed, the metrics requests would be dropped by the MDS as the MDS would still be in the `clientreplay` state. In addition to this, the useless metric requests would slow down MDS. With this fix, the clients need to wait for the MDSs to be ready in the `up:active` state. This allows the clients to send the metrics and avoid slowing down the MDS performance in some cases.
Clone Of: 2224233
Environment:
Last Closed: 2023-08-28 09:40:56 UTC
Embargoed:

Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Ceph Project Bug Tracker 61523 0 None None None 2023-07-20 08:42:48 UTC
Red Hat Issue Tracker RHCEPH-7044 0 None None None 2023-07-20 08:42:00 UTC
Red Hat Product Errata RHBA-2023:4760 0 None None None 2023-08-28 09:41:21 UTC

Description Xiubo Li 2023-07-20 08:41:28 UTC 
+++ This bug was initially created as a clone of Bug #2224233 +++

In some cases when there are a lot of clients and these clients have a lots of known requests need to replay too, the metrics requests will be dropped by the MDS because the MDS is still in the clientreplay state, and also the useless metric requests will slow down MDS.

--- Additional comment from RHEL Program Management on 2023-07-20 08:39:57 UTC ---

Please specify the severity of this bug. Severity is defined here:
https://bugzilla.redhat.com/page.cgi?id=fields.html#bug_severity.
Comment 1 RHEL Program Management 2023-07-20 08:41:43 UTC 
Please specify the severity of this bug. Severity is defined here:
https://bugzilla.redhat.com/page.cgi?id=fields.html#bug_severity.
Comment 15 errata-xmlrpc 2023-08-28 09:40:56 UTC 
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (Red Hat Ceph Storage 5.3 Bug Fix update), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2023:4760
Note You need to log in before you can comment on or make changes to this bug.