[10:50:19] <wm-bb>	 <albertoleoncio> `Failed to connect to MySQL: php_network_getaddresses: getaddrinfo for tools.db.svc.eqiad.wmflabs failed: Temporary failure in name resolution`
[10:51:16] <wm-bb>	 <albertoleoncio> I'm getting this error in toolforge. It's intermittent, but I've never seen it before.
[13:40:15] <bawolff>	 So I'm new to toolforge build service
[13:40:21] <bawolff>	 I'm trying to look at my logs
[13:40:35] <bawolff>	 I've been lead to believe the command for that is toolforge webservice --backend kubernetes buildservice logs -f
[13:41:05] <bawolff>	 however that gives "dateutil.parser._parser.ParserError: String does not contain a date"
[13:41:17] <bawolff>	 in File "/usr/lib/python3/dist-packages/toolforge_weld/logs/kubernetes.py", line 45
[13:41:41] <bawolff>	 Am i doing something wrong here, or is this a bug in toolfroge build service?
[13:43:37] <bawolff>	 Full backtrace: https://dpaste.org/GVQW9/raw
[13:51:14] <bawolff>	 oh, well suddenly my tool works, so maybe there was just a start up delay. So i don't need to see the logs anymore
[13:51:28] <bawolff>	 I'm still not sure if that error message is a log message, or an error saying the logs cannot be read
[13:52:01] <bawolff>	 actually i assume its the latter, since otherwise the -f option would keep things open
[13:52:38] <wm-bb>	 <lucaswerkmeister> it sounds like a bug in the toolforge CLI to me, but I don’t have much experience with the build service either
[13:53:02] <wm-bb>	 <lucaswerkmeister> (btw, I assume you can drop `--backend kubernetes` now that the grid engine is gone)
[13:53:57] <bawolff>	 oh true. However it doesn't make a difference in terms of the error
[13:58:43] <bawolff>	 I wonder if the cause is one of the log messages has a newline in it, so that line doesn't start with a date
[13:58:57] <bawolff>	 wonder if i can access the logs from docker more directly somehow
[14:03:11] <bawolff>	 Oh, i can get around the error by specifying -l argument so the problematic line isn't returned
[14:04:10] <bawolff>	 That solves my immediate problem, but there is still probably a bug here
[14:09:06] <bawolff>	 filed as https://phabricator.wikimedia.org/T373201
[17:44:13] <wm-bb>	 <Arthur> Hi - I'm seeing intermittent errors just like @albertoleoncio showed above, this is in a php kubernetes container: Fatal error: Uncaught mysqli_sql_exception: php_network_getaddresses: getaddrinfo for tools.db.svc.eqiad.wmflabs failed: Temporary failure in name resolution in /data/project/author-disambiguator/public_html/lib/database_tools.php:15 Stack trace: #0
[17:44:13] <wm-bb>	 /data/project/aut <clipped message>
[17:44:15] <wm-bb>	 <Arthur> hor-disambiguator/public_html/lib/database_tools.php(15): mysqli->__construct() #1 /data/project/author-disambiguator/public_html/work_item_oauth.php(7): DatabaseTools->openToolDB() #2 {main} thrown in /data/project/author-disambiguator/public_html/lib/database_tools.php on line 15
[17:45:03] <wm-bb>	 <Arthur> Sometimes it works, sometimes not. I ran 'webservice restart' and same problem continues.
[18:10:24] <wm-bb>	 <Arthur> Ok, seems to have recovered now. Problem with DNS server for kubernetes pods?
[18:59:30] <wm-bb>	 <sohom_datta> CropTool seems to be down since it can't access www.mediawiki.org
[19:04:06] <wm-bb>	 <sohom_datta> I assume it might be related to the problem just above, maybe somebody needs to restart to webservice again ?