| opendevreview | Merged openstack/swift master: reconstructor: restrict max objects per revert job https://review.opendev.org/c/openstack/swift/+/818017 | 02:47 |
|---|---|---|
| opendevreview | Matthew Oliver proposed openstack/swift master: POC: refactored errorlimited with global cache https://review.opendev.org/c/openstack/swift/+/820313 | 09:59 |
| opendevreview | Matthew Oliver proposed openstack/swift master: WIP: Add global error_limiting to object-updater https://review.opendev.org/c/openstack/swift/+/820486 | 09:59 |
| mattoliver | ^ acoles first pass at using a class too | 10:11 |
| acoles | :thu | 10:13 |
| acoles | 👍 | 10:13 |
| opendevreview | Alistair Coles proposed openstack/swift master: internal-client: pass global_conf to loadapp https://review.opendev.org/c/openstack/swift/+/820029 | 11:03 |
| opendevreview | Merged openstack/swift master: reconstructor: include partially reverted handoffs in handoffs_remaining https://review.opendev.org/c/openstack/swift/+/820387 | 14:56 |
| manuvakery1 | Hi what is SW in the source field of following proxy-server log | 15:35 |
| manuvakery1 | Dec 2 10:59:50 swift-proxy2 proxy-server: 99.248.129.100 10.0.0.2 02/Dec/2021/10/59/50 GET /v1/d84a1s87f9bcc84edeb37a619d2f406f/ds9archive/DS920_hyperbackup.hbk/Pool/0/21/511.bucket.536 HTTP/1.0 200 - HyperBackup/2.2.8-1510%20%28DS920%2B%3B%20DSM%206.2-25556%29%20aws-sdk-php2/2.8.31%20Guzzle/3.9.3%20curl/7.54.0%20PHP/5.6.11%20StaticWeb - - 27264116 - tx0121737be1a849e688e10-0061a8a721 - 4.8183 SW - | 15:35 |
| manuvakery1 | 1638442785.342961073 1638442790.161214113 0 | 15:35 |
| opendevreview | Alistair Coles proposed openstack/swift master: Modify internal client logger name https://review.opendev.org/c/openstack/swift/+/820539 | 15:39 |
| acoles | manuvakery1: it's the 'swift source' which indicates the source of an (internally-generated) request, in this case 'Static Web' https://docs.openstack.org/swift/latest/logs.html | 15:42 |
| manuvakery1 | acoles: what do you mean by (internally-generated) request , will there be associated requests when i make a GET call? | 15:44 |
| manuvakery1 | acoles: what will be the source for the actual client request? | 15:48 |
| acoles | some client requests can result in multiple requests internal to Swift, for example with Static Web there may be requests to check container metadata | 15:53 |
| acoles | I don't think that the field is set for the original client request | 15:54 |
| acoles | (I could be wrong on that ^^) | 15:54 |
| manuvakery1 | acoles: thanks for your response. will the transaction id be same for these internally generated requests? | 16:05 |
| acoles | yes, it should be | 16:08 |
| manuvakery1 | in logs my I see 2 requests with the same transaction id , one is GET call an other is a HEAD call .. I assume the HEAD call is internally generated and the GET call is the actual client call. I see in both calls the source is set to SW | 16:18 |
| manuvakery1 | Its critical for me to find the actual bytes send/received from the client | 16:22 |
| manuvakery1 | https://github.com/openstack/ceilometermiddleware/blob/master/ceilometermiddleware/swift.py#L294-L298 | 16:22 |
| manuvakery1 | the ceilometermiddleware here emit events only is the swift.source is None.. what will happen if the calls s3/static web | 16:23 |
| timburke | client request should have a swift.source that's unset; i.e., env.get('swift.source') will be None. in logs, it will show up as a dash: "-" | 16:26 |
| manuvakery1 | timburke which means for any request there will be one line in the log with swift.source is set as "-" | 16:35 |
| manuvakery1 | and more logs with same transaction id but with other value in source field | 16:36 |
| opendevreview | Alistair Coles proposed openstack/swift master: Improve updaters unit tests https://review.opendev.org/c/openstack/swift/+/820579 | 18:09 |
| opendevreview | Tim Burke proposed openstack/swift master: updaters: use replication network https://review.opendev.org/c/openstack/swift/+/820402 | 18:38 |
| zaitcev | You're squashing? I had a question about Alistair's follow-up just now. | 18:38 |
| acoles | zaitcev: what's the question? (no comments on gerrit) | 18:40 |
| zaitcev | Indeed, strange. Anyway, why do you only add replication_ip but not replication_port? https://review.opendev.org/c/openstack/swift/+/820579/1/test/unit/obj/test_updater.py | 18:41 |
| acoles | zaitcev: because the replication_port is set in the test https://review.opendev.org/c/openstack/swift/+/820579/1/test/unit/obj/test_updater.py#542 | 18:43 |
| zaitcev | thanks | 18:44 |
| acoles | but, on reflection, would be no harm setting it in setUp anyway...to avoid us wondering again in the future | 18:44 |
| timburke_ | feel free to push over :-) | 18:44 |
| acoles | will do | 18:48 |
| opendevreview | Alistair Coles proposed openstack/swift master: updaters: use replication network https://review.opendev.org/c/openstack/swift/+/820402 | 18:48 |
| acoles | zaitcev: timburke_ ^^ | 18:49 |
| opendevreview | Tim Burke proposed openstack/swift master: updater: Add timing stats https://review.opendev.org/c/openstack/swift/+/819732 | 19:34 |
| opendevreview | Tim Burke proposed openstack/swift master: updater: Skip updates to overloaded nodes https://review.opendev.org/c/openstack/swift/+/819931 | 19:34 |
| opendevreview | Matthew Oliver proposed openstack/swift master: POC: refactored errorlimited with global cache https://review.opendev.org/c/openstack/swift/+/820313 | 23:59 |
| opendevreview | Matthew Oliver proposed openstack/swift master: WIP: Add global error_limiting to object-updater https://review.opendev.org/c/openstack/swift/+/820486 | 23:59 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!