Aborted connection 29 to db: 'unconnected' user: 'root' host: 'cloudsqlproxy~xxx.xxx.xx.xx' (Got an error reading communication packets)

[Peter Riisager]

Please if anybody have insights or suggestions for the aborted connection error we are experiencing  on our Cloud SQL. The database is operational, but every every 2-3 minutes we are getting errors similar to this:

2017-05-25T12:36:30.341228Z 29 [Note] Aborted connection 29 to db: 'unconnected' user: 'root' host: 'cloudsqlproxy~xxx.xxx.xx.xx' (Got an error reading communication packets)

We believe it is related to our MySQL instance. The error occurs when our App Engine application reads from the Cloud SQL. It can also occur from an MySQL Workbench connection the database.

Don't know if this is related but on restarting the Cloud SQL we get the following list of Errors:

E  + rm -f /mysql/mysql.sock.lock 
E  + cat /mysql/my.cnf 
E  + exec /usr/sbin/mysqld --defaults-file=/mysql/my.cnf 
E  2017-05-25T12:35:23.168460Z 0 [Warning] TIMESTAMP with implicit DEFAULT value is deprecated. Please use --explicit_defaults_for_timestamp server option (see documentation for more details). 
E  2017-05-25T12:35:23.168501Z 0 [Warning] Insecure configuration for --secure-file-priv: Current value does not restrict location of generated files. Consider setting it to a valid, non-empty path. 
E  2017-05-25T12:35:23.168531Z 0 [Note] /usr/sbin/mysqld (mysqld 5.7.14-google-log) starting as process 1 ... 
E  2017-05-25T12:35:23.171393Z 0 [Note] Plugin 'AUDIT_LOG' is disabled. 
E  2017-05-25T12:35:23.171422Z 0 [Note] Plugin 'FEDERATED' is disabled. 
E  2017-05-25T12:35:23.171899Z 0 [Warning] InnoDB: Using innodb_file_format is deprecated and the parameter may be removed in future releases. See http://dev.mysql.com/doc/refman/5.7/en/innodb-file-format.html 
E  2017-05-25T12:35:23.171931Z 0 [Note] InnoDB: PUNCH HOLE support not available 
E  2017-05-25T12:35:23.171938Z 0 [Note] InnoDB: Mutexes and rw_locks use GCC atomic builtins 
E  2017-05-25T12:35:23.171943Z 0 [Note] InnoDB: Uses event mutexes 
E  2017-05-25T12:35:23.171947Z 0 [Note] InnoDB: GCC builtin __atomic_thread_fence() is used for memory barrier 
E  2017-05-25T12:35:23.171951Z 0 [Note] InnoDB: Compressed tables use zlib 1.2.8 
E  2017-05-25T12:35:23.171955Z 0 [Note] InnoDB: Using Linux native AIO 
E  2017-05-25T12:35:23.177248Z 0 [Note] InnoDB: Number of pools: 1 
E  2017-05-25T12:35:23.177411Z 0 [Note] InnoDB: Using CPU crc32 instructions 
E  2017-05-25T12:35:23.180087Z 0 [Note] InnoDB: Initializing buffer pool, total size = 11G, instances = 8, chunk size = 128M 
E  2017-05-25T12:35:23.884265Z 0 [Note] InnoDB: Completed initialization of buffer pool 
E  2017-05-25T12:35:24.072950Z 0 [Note] InnoDB: If the mysqld execution user is authorized, page cleaner thread priority can be changed. See the man page of setpriority(). 
E  2017-05-25T12:35:24.096177Z 0 [Note] InnoDB: Highest supported file format is Barracuda. 
E  2017-05-25T12:35:24.441336Z 0 [Note] InnoDB: Removed temporary tablespace data file: "ibtmp1" 
E  2017-05-25T12:35:24.441364Z 0 [Note] InnoDB: Creating shared tablespace for temporary tables 
E  2017-05-25T12:35:24.441425Z 0 [Note] InnoDB: Setting file './ibtmp1' size to 12 MB. Physically writing the file full; Please wait ... 
E  2017-05-25T12:35:24.502928Z 0 [Note] InnoDB: File './ibtmp1' size is now 12 MB. 
E  2017-05-25T12:35:24.504142Z 0 [Note] InnoDB: 96 redo rollback segment(s) found. 96 redo rollback segment(s) are active. 
E  2017-05-25T12:35:24.504163Z 0 [Note] InnoDB: 32 non-redo rollback segment(s) are active. 
E  2017-05-25T12:35:24.504540Z 0 [Note] InnoDB: Waiting for purge to start 
E  2017-05-25T12:35:24.556628Z 0 [Note] InnoDB: 5.7.14 started; log sequence number 64073224309 
E  2017-05-25T12:35:24.557111Z 0 [Note] InnoDB: Loading buffer pool(s) from /mysql/datadir/ib_buffer_pool 
E  2017-05-25T12:35:24.562549Z 0 [Note] Semi-sync replication initialized for transactions. 
E  2017-05-25T12:35:24.562584Z 0 [Note] Semi-sync replication enabled on the master. 
E  2017-05-25T12:35:24.686132Z 0 [Note] InnoDB: Buffer pool(s) load completed at 170525 12:35:24 
E  2017-05-25T12:35:24.725836Z 0 [Note] Skip crash recovery, binlog file '/mysql/binlog/mysql-bin.000175' size: 217 
E  2017-05-25T12:35:24.726796Z 0 [Note] Starting ack receiver thread 
E  2017-05-25T12:35:24.741596Z 0 [Note] Found gtids in binlog: 'd787603e-6b86-11e6-bcc8-42010af00d90:1-20355638' 
E  2017-05-25T12:35:24.741672Z 0 [Note] Found gtids purged from binlog: 'd787603e-6b86-11e6-bcc8-42010af00d90:1-19649884' 
E  2017-05-25T12:35:24.746277Z 0 [Note] Skipping generation of SSL certificates as options related to SSL are specified. 
E  2017-05-25T12:35:24.748078Z 0 [Warning] CA certificate /mysql/datadir/client_ca_cert.pem is self signed. 
E  2017-05-25T12:35:24.748153Z 0 [Note] Skipping generation of RSA key pair as key files are present in data directory. 
E  2017-05-25T12:35:24.748800Z 0 [Note] Server hostname (bind-address): '*'; port: 3306 
E  2017-05-25T12:35:24.748865Z 0 [Note] IPv6 is available. 
E  2017-05-25T12:35:24.748885Z 0 [Note]   - '::' resolves to '::'; 
E  2017-05-25T12:35:24.748899Z 0 [Note] Server socket created on IP: '::'. 
E  2017-05-25T12:35:24.758841Z 0 [Warning] 'user' entry 'root@localhost' ignored in --skip-name-resolve mode. 
E  2017-05-25T12:35:24.758891Z 0 [Warning] 'user' entry 'cloudsqlimport@localhost' ignored in --skip-name-resolve mode. 
E  2017-05-25T12:35:24.758902Z 0 [Note] Loaded users from system_user table: root@localhost,cloudsqlreplica@%,cloudsqlimport@localhost. 
E  2017-05-25T12:35:24.758945Z 0 [Warning] 'user' entry 'mysql.sys@localhost' ignored in --skip-name-resolve mode. 
E  2017-05-25T12:35:24.758967Z 0 [Warning] 'db' entry 'sys mysql.sys@localhost' ignored in --skip-name-resolve mode. 
E  2017-05-25T12:35:24.795794Z 0 [Warning] 'tables_priv' entry 'sys_config mysql.sys@localhost' ignored in --skip-name-resolve mode. 
E  2017-05-25T12:35:24.813306Z 0 [Note] Event Scheduler: Loaded 0 events 
E  2017-05-25T12:35:24.813511Z 0 [Note] /usr/sbin/mysqld: ready for connections. 
E  Version: '5.7.14-google-log'  socket: '/mysql/mysql.sock'  port: 3306  (Google) 
E  2017-05-25T12:35:28.705163Z 10 [Warning] Timeout waiting for reply of binlog (file: mysql-bin.000176, pos: 521), semi-sync up to file , position 0. 
E  2017-05-25T12:35:28.705196Z 10 [Note] Semi-sync replication switched OFF. 

[George (Cloud Platform Support)]

Hello Peter, 

This type of errors is usually caused by networking and connectivity issues, such as badly configured TCP/IP, faulty Ethernets, hubs, switches, cables. For diagnosing purposes it may be worthwhile trying to connect from a different network and check for any differences in eventual error messages. 

If connectivity is not an issue, your can try raising the max_allowed_packet to a  larger number, upwards from the default value of 1M. 

For more detail, have a look at the "B.5.2.11 Communication Errors and Aborted Connections" online document. 

If you have any more information facilitating our reproduction of the issue, you are encouraged to add it to your reply. 

[Danqing Liu]

We are having exactly the same issue, from Compute Engine.

George, this does not seem to be a networking/connectivity issue since this is GCP internal. Any other idea why these things occur? They occur like 20+ times a day - not rare at all.

Thanks!