Increased rate of "MySQL server has gone away" errors

[geoffre...@gmail.com]

Hello,

As of today we're seeing an increased rate of "MySQL server has gone away" errors when connecting to our Cloud SQL 2G instance. Prior to today we might see this error once every few months, but today we've seen it 6 times according to App Engine Error Reporting and 18 times according to our App Engine logs. 

I do not see an update operation on our instance, so any thoughts on where to start looking would be greatly appreciated.

Best,

Geoff.

[George (Cloud Platform Support)]

Yours is an intermittent issue, one of the sort that is notoriously difficult to debug. There is an excellent page “B.5.2.9 MySQL server has gone away” to start from, in the MySQL documentation itself.

Intermittent occurrences of the same kind have been reported by means of the “random ‘server has gone away’ errors from GAE to cloud SQL” public issue tracker entry. You may consider reporting your case there. This may speed up the research for an eventual resolution or workaround.

Have you checked the values of your max_allowed_packet or wait_timeout parameters?

For a more in-depth look at this situation, details such as your instance ID and project name are necessary. You can send this info in confidential manner through a private email.

[Geoffrey Arnold]

Hey George, I actually already added a +1 to that issue.

We haven't seen a "server has gone away" since I reported the issue on Monday. Monday definitely seems like something else was going on. Were there any issues on GCP that day?

--
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+unsub...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/google-cloud-sql-discuss/02f6ed3f-596a-42b7-ac3b-d5ed5d5f0a96%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[George (Cloud Platform Support)]

Hello Geoffrey!

You are right, there was a similar issue on that day, but of the same random and non-reproducible kind. I cannot say it is related, as I have no information as to the region and zone of your project. It has been merged in the issue linked above, where you had posted your comment already. You may consider keeping track of that issue, to get updates. 

Hoping this is of help now and that it will offer soon a resolution to your problem.  

[Geoffrey Arnold]

Thanks George.

On Wed, Mar 1, 2017 at 11:09 AM, 'George (Cloud Platform Support)' via Google Cloud SQL discuss <google-cloud...@googlegroups.com> wrote:

--

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+unsub...@googlegroups.com.

[Tech Support]

Hi, 

Even we have the same issue. Every Monday I see this error "MySQL error has gone away". I have set the parameters sufficient enough "wait_time" and 'allowed_packet _size'. No idea why this is happening.

[Olu]

Hello, 

Have you gone through the Issue link[1] referenced by George in one of Messages? I recommend that you do that as this issue may be related. 

That being said, I believe there are a number of reasons that may be responsible for this type of errors. For example, I gathered that this error message may be reported following errors about attempts at reading communication packets. 

Of course, when you have errors about communication packets, it often suggests that an existing connection is perhaps terminated improperly. Aborted connections happen because of unclean’s connection closure, or a networking issue between the Cloud SQL MySQL instance and the client.

Whether or not this is the situation on your Case would require reviewing your Project/setup and the logs to understand what led to the errors, which is why I recommend that you contact the GCP Support Engineers[2] so that the team could help with determining the cause of the error particularly on your CloudSQL Instance. 

[1]https://issuetracker.google.com/35904221

[2]https://cloud.google.com/support-hub