[google-cloud-sql-discuss] Re: After Google Maintenance CloudSQL stopped working

We were able to replicate the issue on a local MySQL instance.  There was a query being run that was causing a pointer exception.  We resolved the issue by changing the query.  It's possible that one of the MySQL libraries that Google updated has a bug.  This code had been working in CloudSQL before.

On Monday, February 24, 2020 at 10:08:49 AM UTC-9, Devin Homan wrote:

I am having intermittent connection issues since yesterday's maintenance.  It looks like it might be an internal problem with the mysqld.  Perhaps the maintenance created bad linked libraries.

key_buffer_size = 8388608

read_buffer_size = 131072

sort_buffer_size = 262144

max_threads = 4030

key_buffer_size + (read_buffer_size + sort_buffer_size) * max_threads = 1609833 K bytes of memory.

The log suggests that one of the linked libraries may be corrupt. I  see

Thread pointer: 0x34b20d125000

Cannot determine thread, fp=7f71fabfe770

On Monday, February 24, 2020 at 7:23:22 AM UTC-9, Hans Ravnaas wrote:

Arvind, I'm seeing SQL issues as well after the maintenance, so posting here in case it's related:

We are using Java 11 and Postgres 11. Our code has not changed in the last two weeks, but after the maintenance on 2/23, we starting having this error in the SQL server logs:

LOG: SSL error: DATA_LENGTH_TOO_LONG

Any chance this is what you are seeing too?

Hans

On Sunday, February 23, 2020 at 12:08:59 AM UTC-8, Arvind Stutzen wrote:

Log(attached below) shows that after Google updated the CloudSQL version this problem started. Still no response from Tech Support for created P1 case and our business been seriously affected by this 13+hrs downtime :(  . Any suggestions how it can be solved?

On Sunday, February 23, 2020 at 12:02:26 PM UTC+5:30, Arvind Stutzen wrote:

After Google Maintenance CloudSQL stopped working.  In error logs it displays as the server keeps restarting. 

Also in parallel tried to create a new instance and restore the old automated backup, even that too fails in the middle of restoring  :( . 

Using paid support subscription created ticket but it is not addressed until now. cant able to accept that Production DB down for 12hours. Being SaaS DB numerous customers are affected by it.  Are there any suggestions to get it resolved? 

--
You received this message because you are subscribed to the Google Groups "Google Cloud SQL discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to google-cloud-sql-discuss+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/google-cloud-sql-discuss/4635fb34-2a1a-425a-9fc8-e71728c8b3c1%40googlegroups.com.