To add to this I’ve noticed that for RGW with s3website api enabled it’s even worse - sometimes it needs a hard reset. Tested on 17.2.6.
Ondra
> On 26. 8. 2024, at 4:48, Huy Nguyen <viplanghe6@xxxxxxxxx> wrote:
>
> Hi community,
> I'm using ceph v18.2.4. Each time I commit a period, all of my radosgw instances paused for about 30s
> which is kinda long, and make my object service down during that time.
>
> how to avoid the down time? Is there better/safer practice?
>
> Thanks
>
> Here is radosgw logs each time I did a period commit. You can see it was paused for a while since the last request
>
> 2024-08-23T22:43:32.152+0700 7f4594d2f700 1 ====== starting new request req=0x7f42dcfbf730 =====
> 2024-08-23T22:43:32.152+0700 7f438b11c700 1 ====== req done req=0x7f42dcfbf730 op status=0 http_status=200 latency=0.000000000s ======
> 2024-08-23T22:43:32.152+0700 7f438b11c700 1 beast: 0x7f42dcfbf730: 172.24.9.101 - anonymous [23/Aug/2024:22:43:32.152 +0700] "GET / HTTP/1.0" 200 231 - "check_http/v2.2 (monitoring-plugins 2.2)" - latency=0.000000000s
> 2024-08-23T22:44:02.778+0700 7f42e1fca700 1 rgw realm reloader: Pausing frontends for realm update...
> 2024-08-23T22:44:02.778+0700 7f42e1fca700 1 rgw realm reloader: Frontends paused
> 2024-08-23T22:44:03.358+0700 7f42e1fca700 1 rgw realm reloader: driver closed
> 2024-08-23T22:44:03.497+0700 7f42e1fca700 1 rgw realm reloader: Creating new driver
> 2024-08-23T22:44:03.497+0700 7f42e1fca700 1 mgrc service_daemon_register rgw.5055090799 metadata {arch=x86_64,ceph_release=reef,ceph_version=ceph version 18.2.4 (e7ad5345525c7aa95470c26863873b581076945d) reef (stable),ceph_version_short=18.2.4....
> 2024-08-23T22:44:03.497+0700 7f42e1fca700 1 rgw realm reloader: Finishing initialization of new driver
> 2024-08-23T22:44:03.497+0700 7f42e1fca700 1 rgw realm reloader: - REST subsystem init
> 2024-08-23T22:44:03.497+0700 7f42e1fca700 1 rgw realm reloader: - usage subsystem init
> 2024-08-23T22:44:03.497+0700 7f42e1fca700 1 rgw realm reloader: Resuming frontends with new realm configuration.
> 2024-08-23T22:44:06.135+0700 7f4637674700 1 ====== starting new request req=0x7f42dcfbf730 =====
> 2024-08-23T22:44:06.136+0700 7f4637674700 -1 res_query() failed
> 2024-08-23T22:44:06.136+0700 7f4637674700 1 ====== req done req=0x7f42dcfbf730 op status=0 http_status=405 latency=0.000999984s ======
> 2024-08-23T22:44:06.136+0700 7f4637674700 1 beast: 0x7f42dcfbf730: 172.24.17.21 - - [23/Aug/2024:22:44:06.135 +0700] "GET /admin HTTP/1.1" 405 133 - "curl/7.61.1" - latency=0.000999984s
> _______________________________________________
> ceph-users mailing list -- ceph-users@xxxxxxx
> To unsubscribe send an email to ceph-users-leave@xxxxxxx
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
