MS is down. The appropriate people have been pinged. SD functionality is degraded. Changes made to the watchlists and blacklists should be stored on GitHub, but will not automatically become functional. Once CI has passed, someone with write access to the SD repo will need to manually move the HEAD of the deploy branch to match the master branch. The only SD instance currently operational is Makyen/Windows. If there are problems with the instance, please ping me.
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 03:58:41 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/3632615
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 03:58:55 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/3631995
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 03:59:09 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/57898
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 03:59:24 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/3632622
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 03:59:39 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/220274
I guess MS only makes the deploy branch change for commits not made by SD. So, !!/watch and !!/blacklist-* commands should work normally. It looks like the problem is only for manual commits.
When MS recovers. Be sure to switch away from using my instance, as it is currently configured not to use MS, so even if MS comes back online my instance won't use it.
[ SmokeDetector ] Url-only title, bad ip for hostname in body, bad ip for hostname in title, bad ns for domain in body, bad ns for domain in title, +6 more (1086): www.reviewstore.co.uk/provia-max/ by mexdenifer4 on askubuntu.com
@Makyen That pattern looks like it's already caught by Pattern-matching website in answer and Pattern-matching website in body; append -force if you really want to do that.
@Makyen That pattern looks like it's already caught by Potentially bad keyword in answer and Potentially bad keyword in body; append -force if you really want to do that.
@Makyen That pattern looks like it's already caught by Potentially bad keyword in answer and Potentially bad keyword in body; append -force if you really want to do that.
OK. My SD instance is now back to reporting to MS. If you authorize your browser to use metasmoke's expired SSL certificate, you should be able to use MS, AIM, FIRE, etc. normally.
Unfortunately, the error which the other SD instances are stuck on happens before any opportunity to command them to pull a new version (and they would only be able to get chat commands, not pull commands from MS, due to the SSL issues with MS). As a result, to get the other instances working, either the instance owners will need to pull the most recent version of master and deploy, or we'll need metasmoke's SSL certificate fixed.
@Makyen Trying to get the certificate working would be the best, as github probably won't be able to send the status of pushed commits and finished builds to MS because of the certificate problem, meaning things like autopull won't work
@Ferrybig Yes, renewing the SSL certificate is certainly the preferred solution. However, that requires @ThomasWard. Getting the other normally active SD instances up requires either @ThomasWard or @iBug. Alternately, if my instance goes down, it's now possible for one of the other SD runners to bring up an instance using a normal config file, rather than having to disable MS in the config file.
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 13:40:05 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/545036
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 13:40:20 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/203621
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 13:40:34 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/357929
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 13:40:49 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/63377
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 13:41:03 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/3633145
@Kulfy That pattern looks like it's already caught by Potentially bad keyword in answer and Potentially bad keyword in body; append -force if you really want to do that.
@Makyen so about the certs looks like LE altered things and now only lets us HTTPS auth which... causes some strife. May have to ‘down’ MS briefly to do the cert auth...
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 14:05:16 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/3633172
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 14:05:31 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/87715
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 14:05:45 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/27963
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 14:05:59 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/3632660
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 14:06:14 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/160501
The quota_remaining property was not in the API response. Error: Violation of backoff parameter at 14:06:29 UTC. Backing off on requests for the next 12 seconds. Previous URL: https://api.stackexchange.com/2.2/questions/161041
I've back-filled MS with scanning the posts which were missed. Obviously, that won't include any which had been deleted. It also doesn't include a few which have been edited since SD initially detected them.
@Undo typically, if you want to have a subdomain with separate authoritative DNS components you'd define a separate NS record and point to a FQDN of a DNS server that's authoritative for that subdomain/section
but that essentially hands off all control for that subdomain portion to another DNS server
that's the only way I know of to get separate A/TXT records work globally
the problem being that the TXT fields don't automatically support includes :/
@ThomasWard At least my SD instance was having an issue completing update_tld_names(). It didn't appear to be actually required, so instead of raising an error, it now logs a warning, if update_tld_names() raises a TldIOError. It should be looked at in more detail to see what was really going on. The other thing which was necessary to have SD work was to tell it that it didn't need to verify the SSL certificate for MS.
example.com is on Google DNS, but sub.example.com has its own authoritative Nameservers at dns.googleplex.example. DNS entries for example.com need an NS record added as follows:
`sub.example.com IN NS dns.google.plex.example.`
This says to all DNS servers "I am authoritative for example.com, but sub.example.com is actually over on this other server."
that would delegate all DNS data for sub.example.com to the other nameserver - A, CNAME, AAAA, TXT, etc.
@ThomasWard Yeah, it's a call to the tld plugin. I intend to take a better look at it now that the SSL certificate for MS is working. I've been assuming that the issue with update_tld_names() is not actually related to the SSL issue on MS. My change there was intended to just work around the problem until I had the time to take a look at what was going on.
