Fwd: Could SQL instance hung at backup

52 views
Skip to first unread message

Faizan (Google Cloud Support)

unread,
May 16, 2016, 7:23:20 PM5/16/16
to google-cloud...@googlegroups.com, nickge...@gmail.com
Hello Nicholas,

This question is related to Google CloudSQL as such, I am forwarding this thread to CloudSQL discussion group. Further followups on this thread can be done on that group.

I hope that helps.

Faizan

Faizan

unread,
May 16, 2016, 7:54:05 PM5/16/16
to Google Cloud SQL discuss, nickge...@gmail.com
I have tried to forward a thread from GCE discussion group to this group but for some reason the original question is not showing here. For the reference Nicholas question is posted on this thread.

Faizan

John Kanalakis

unread,
May 17, 2016, 12:23:30 AM5/17/16
to Google Cloud SQL discuss, nickge...@gmail.com
I have the same problem with a Cloud SQL database in us-central (advertopia-mysql-2). The control panel first indicated "Running Backup..." and blocked all other options (import, export, restart, stop, delete). Now, that message is no longer appearing and the database is dead. I am not able to SSH into it to restart (connection eventually times out) and I am completely stuck. My Website is dead. I think somthing went wrong in the datacenter during the backup and the instance still thinks (incorrectly) that the backup is in progress.

I am trying without much luck to force a restart from curl with this command:

curl -X POST -d "" https://www.googleapis.com/sql/v1beta4/projects/Advertopia/instances/advertopia-mysql-2/restart


...but now I'm struggling with authorization...


{

 "error": {

  "errors": [

   {

    "domain": "global",

    "reason": "required",

    "message": "Login Required",

    "locationType": "header",

    "location": "Authorization"

   }

  ],

  "code": 401,

  "message": "Login Required"

 }

}


Any ideas or advice is greatly appreciated.

John K.

Nicholas Geovanis

unread,
May 17, 2016, 10:46:26 AM5/17/16
to Google Cloud SQL discuss, nickge...@gmail.com
In my case, simply waiting from about 9:30pm until subsequent 8:30am gave it enough time to recover (total "recovery" time was about 20 hours). Apparently it's in good condition. It is only a 10GB instance so one thinks that was not really a recovery, something else going on.

Tianzhou Chen

unread,
May 17, 2016, 5:14:08 PM5/17/16
to Google Cloud SQL discuss, nickge...@gmail.com
There is a service upgrade starting from Saturday causing some disruption. All the affected instances should be fixed now. Please let us know if you still experience issue.

Sorry for the inconvenience.

Tianzhou
Reply all
Reply to author
Forward
0 new messages