Failure to calculate super-latch promotion threshold

  • I saw this error message in the error log of one of my SQL Server 2008 R2 servers today (with SP2 installed, v10.50.4000).

    "Warning: Failure to calculate super-latch promotion threshold."

    I've never seen that message, can't find anything online about it, and don't know what the impact is.

    The only other relevant information I can give is that there was a series of DBCC CHECKDB commands occurring at the time (our nightly database checks).

    Any additional information would be appreciated.

    Thanks,

    John

  • what is the error number, state and severity from the error log?

  • None, I copied and pasted the message exactly as it was in the log.

  • Nothing directly underneath that line, which looks something like

    Error: ####, Serverity: ##, State: ##.

    Usually they span two lines

    Login failed for user 'someuser'. Reason: Failed to open the explicitly specified database. [CLIENT: 192.168.0.1]

    Error: 18456, Severity: 14, State: 38.

  • This is exactly how it came in (with the exception of me renaming actual application DBs to DB1, DB2, etc...)

    2012-09-18 00:00:47.870 spid20s This instance of SQL Server has been using a process ID of 15280 since 8/7/2012 2:08:02 PM (local) 8/7/2012 6:08:02 PM (UTC). This is an informational message only; no user action is required.

    2012-09-18 01:00:01.990 spid59 DBCC CHECKDB (master) WITH all_errormsgs, no_infomsgs executed by PROD\devsvc found 0 errors and repaired 0 errors. Elapsed time: 0 hours 0 minutes 1 seconds. Internal database snapshot has split point LSN = 0000029a:0000003c:0002 and first LSN = 0000029a:0000003b:0001.

    2012-09-18 01:00:12.880 spid59 DBCC CHECKDB (mssqlsystemresource) WITH all_errormsgs, no_infomsgs executed by PROD\devsvc found 0 errors and repaired 0 errors. Elapsed time: 0 hours 0 minutes 10 seconds.

    2012-09-18 01:00:15.450 spid59 DBCC CHECKDB (model) WITH all_errormsgs, no_infomsgs executed by PROD\devsvc found 0 errors and repaired 0 errors. Elapsed time: 0 hours 0 minutes 2 seconds. Internal database snapshot has split point LSN = 00000016:00000085:0001 and first LSN = 00000016:00000084:0001.

    2012-09-18 01:00:40.880 spid59 DBCC CHECKDB (msdb) WITH all_errormsgs, no_infomsgs executed by PROD\devsvc found 0 errors and repaired 0 errors. Elapsed time: 0 hours 0 minutes 25 seconds. Internal database snapshot has split point LSN = 0000a065:000000dc:0001 and first LSN = 0000a065:000000db:0001.

    2012-09-18 01:00:42.240 spid59 DBCC CHECKDB (ReportServer) WITH all_errormsgs, no_infomsgs executed by PROD\devsvc found 0 errors and repaired 0 errors. Elapsed time: 0 hours 0 minutes 1 seconds. Internal database snapshot has split point LSN = 0000003a:00000165:0001 and first LSN = 0000003a:00000164:0001.

    2012-09-18 01:00:42.770 spid59 DBCC CHECKDB (ReportServerTempDB) WITH all_errormsgs, no_infomsgs executed by PROD\devsvc found 0 errors and repaired 0 errors. Elapsed time: 0 hours 0 minutes 0 seconds. Internal database snapshot has split point LSN = 00000018:0000010f:0001 and first LSN = 00000018:0000010e:0001.

    2012-09-18 02:07:39.340 spid4s Warning: Failure to calculate super-latch promotion threshold.

    2012-09-18 02:11:06.130 spid59 DBCC CHECKDB (DB1) WITH all_errormsgs, no_infomsgs executed by PROD\devsvc found 0 errors and repaired 0 errors. Elapsed time: 1 hours 10 minutes 23 seconds. Internal database snapshot has split point LSN = 00199f9b:000000c8:0001 and first LSN = 00199f9b:000000c7:0001.

    2012-09-18 02:46:22.320 spid59 DBCC CHECKDB (DB2) WITH all_errormsgs, no_infomsgs executed by PROD\devsvc found 0 errors and repaired 0 errors. Elapsed time: 0 hours 35 minutes 13 seconds. Internal database snapshot has split point LSN = 0032527a:0000102f:0001 and first LSN = 0032527a:0000102e:0001.

    2012-09-18 02:48:16.720 spid59 DBCC CHECKDB (DB3) WITH all_errormsgs, no_infomsgs executed by PROD\devsvc found 0 errors and repaired 0 errors. Elapsed time: 0 hours 1 minutes 53 seconds. Internal database snapshot has split point LSN = 0005e0b7:000003f8:0001 and first LSN = 0005e0b7:000003f7:0001.

    2012-09-18 03:20:06.140 spid59 DBCC CHECKDB (DB4) WITH all_errormsgs, no_infomsgs executed by PROD\devsvc found 0 errors and repaired 0 errors. Elapsed time: 0 hours 31 minutes 48 seconds. Internal database snapshot has split point LSN = 001422e6:0000c241:0001 and first LSN = 001422e6:0000c240:0001.

    2012-09-18 03:56:48.510 spid59 DBCC CHECKDB (DB5) WITH all_errormsgs, no_infomsgs executed by PROD\devsvc found 0 errors and repaired 0 errors. Elapsed time: 0 hours 36 minutes 41 seconds. Internal database snapshot has split point LSN = 000add1e:000022a1:0001 and first LSN = 000add1e:000022a0:0001.

    2012-09-18 03:58:21.960 spid59 DBCC CHECKDB (DB6) WITH all_errormsgs, no_infomsgs executed by PROD\devsvc found 0 errors and repaired 0 errors. Elapsed time: 0 hours 1 minutes 32 seconds. Internal database snapshot has split point LSN = 0047c74c:000000fd:0001 and first LSN = 0047c74c:000000fc:0001.

    2012-09-18 04:08:54.060 spid59 DBCC CHECKDB (DB7) WITH all_errormsgs, no_infomsgs executed by PROD\devsvc found 0 errors and repaired 0 errors. Elapsed time: 0 hours 10 minutes 31 seconds. Internal database snapshot has split point LSN = 0031884b:000000d6:0001 and first LSN = 0031884b:000000d5:0001.

    2012-09-18 04:08:55.510 spid59 DBCC CHECKDB (DB8) WITH all_errormsgs, no_infomsgs executed by PROD\devsvc found 0 errors and repaired 0 errors. Elapsed time: 0 hours 0 minutes 0 seconds. Internal database snapshot has split point LSN = 0000005d:0000017f:0001 and first LSN = 0000005d:0000017e:0001.

  • Thanks for that, I was thinking it was a user invoked error message as I couldnt find an article on it after a Google.

    One for some reading up on I think, found this on how it works but nothing yet relating to the error

    http://blogs.msdn.com/b/psssql/archive/2009/01/28/hot-it-works-sql-server-superlatch-ing-sub-latches.aspx

  • Thanks for that. I guess I can chalk this up to some kind of disk latency issue?

  • Given that there's not a single occurrence of this error referenced on Google, I'd assume you're hitting a very unusual condition. It sounds relatively innocuous (the error log does record some irrelevant warnings at times), but you might want to log a call with MS Support to get some more information if it's an important production server.

  • I have this warning about 10 times since I installed SP2 on SQL 2008 R2 on October 21st - I am investigating the problem.

  • We also started getting this error immediately after installing SP2 on our 2008 R2 instance.

  • I just checked all the servers that we've received the error from and here are the version numbers...

    SQL 2012 - 11.0.3000

    SQL 2008 R2 - 10.50.4000

  • Any updates to this thread? We just received this error on a SQL 2008 R2 server here in our environment.

  • I've has this message 4 times in the past month on a SQL2008 R2 server. No performance issues so it seems pretty harmless.

  • Error msg - Failure to calculate super-latch promotion threshold - also known as Sub Latch. - Supporting article posted back in 09

    http://blogs.msdn.com/b/psssql/archive/2010/03/24/how-it-works-bob-dorr-s-sql-server-i-o-presentation.aspx

    Sub-latch is also referred to as super latch. These are only used for data cache page latches. They are designed to reduce the latching contention on hot pages. For example if you have a lookup table that is only a few pages in size but used by 100s of queries per second that SH latch activity is aggressive to protect the page. When SQL Server detects high rates of SH latch activity for a sustained period a buffer latch is PROMOTED to a sub-latch. A sub-latch partitions a single latch into an array of latch structures per logical CPU. In doing so the worker (always assigned to a specific CPU) only needs to acquire a SH on the sub-latch assigned to the local scheduler. This avoids interlocked activity and cache line invalidations across all physical CPUs. The acquiring of an SH sub-latch resource uses less resources and scales access to hot pages better.

  • I just got this error message also on one of my servers. It is also 2008 R2 with SP2, but has had SP2 for a while now.

Viewing 15 posts - 1 through 14 (of 14 total)

You must be logged in to reply to this topic. Login to reply