Bug #44183 [MgmSrvr] INFO Node X: Data usage increased to 100% (msg overload)

• Submitted: 9 Apr 2009 12:26
• Modified: 3 Mar 2010 12:15
• Reporter: Jonathan Miller
• Status: Closed
• Category: MySQL Cluster: Cluster (NDB) storage engine
• Severity: S3 (Non-critical)
• Version: mysql-5.1-telco-7.0
• OS: Any
• Assigned to: Magnus Blåudd
• CPU Architecture: Any

[9 Apr 2009 12:26] Jonathan Miller

Description:
Hi,

I started a test run right before going to bed last night. I awoke to find GB and GB of the following message:

2009-04-09 13:02:46 [MgmSrvr] INFO     -- Node 2: Data usage increased to 100%(2560 32K pages of total 2560)
2009-04-09 13:02:46 [MgmSrvr] INFO     -- Node 3: Data usage increased to 100%(2560 32K pages of total 2560)
2009-04-09 13:02:47 [MgmSrvr] INFO     -- Node 2: Data usage increased to 100%(2560 32K pages of total 2560)
2009-04-09 13:02:47 [MgmSrvr] INFO     -- Node 3: Data usage increased to 100%(2560 32K pages of total 2560)
2009-04-09 13:02:48 [MgmSrvr] INFO     -- Node 2: Data usage increased to 100%(2560 32K pages of total 2560)
2009-04-09 13:02:48 [MgmSrvr] INFO     -- Node 3: Data usage increased to 100%(2560 32K pages of total 2560)

The message is printed every second for all nodes. 

How to repeat:
run a cluster out of data memory

Suggested fix:
print message once for all datanodes, and then print an additional message of action needed. 

i.e. in this case it would be to increase data memory, or reduce data in database.

[9 Apr 2009 13:27] Jonathan Miller

PS: Just as a side test, I took down the cluster and restarted. It *started*
w/o issue and then continued the message every second.

[3 Mar 2010 10:30] Magnus Blåudd

Fixed by patch for BUG#31542

Pushed to 6.3.33, 7.0.14 and 7.1.2-beta

[3 Mar 2010 12:15] Jon Stephens

See BUG#31542 for changelog entry.

Closed.