| Title: | Microsoft SQL Server Support |
| Notice: | Please Registar, Note #11 |
| Moderator: | AMCUCS::BETTS |
| Created: | Tue Aug 23 1994 |
| Last Modified: | Wed Jun 04 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 382 |
| Total number of notes: | 1233 |
Hi, I have a customer who is running SQL 6.5 SP2 on NT3.51 SP5 (Post SP hotfixes also applied) on a 4100. The AlphaBios is on 5.28 and is using REV. C of the NT 3.51 HAL. Customer runs some custom ODBC client app. to talk to SQL. The customer is experiencing problems on the machine where SQL Server dies. (See attached SQL log) According to the customer it happens randomly and there is no specific incident that triggers the crash. The OS runs without any problems when this happens. Herewith the SQL log: 97/05/19 20:03:17.19 kernel Microsoft SQL Server 6.50 - 6.50.240 (ALPHA) Dec 17 1996 16:18:43 Copyright (c) 1988-1996 Microsoft Corporation 97/05/19 20:03:17.19 kernel Copyright (C) 1988-1994 Microsoft Corporation. 97/05/19 20:03:17.19 kernel All rights reserved. 97/05/19 20:03:17.19 kernel Logging SQL Server messages in file 'D:\MSSQL65\LOG\ERRORLOG' 97/05/19 20:03:17.28 kernel initconfig: number of user connections limited to 200 97/05/19 20:03:17.28 kernel SQL Server is starting at priority class 'normal' with dataserver semaphore = 1 (1 CPU detected, SMPStat=1). 97/05/19 20:03:17.31 server LogLRU buffers: 0 97/05/19 20:03:17.48 kernel Attempting to initialize Distributed Transaction Coordinator 97/05/19 20:03:21.13 kernel initializing virtual device 0, D:\MSSQL65\DATA\MASTER.DAT 97/05/19 20:03:21.13 kernel Opening Master Database ... 97/05/19 20:03:21.17 spid1 Loading SQL Server's default sort order and character set 97/05/19 20:03:21.20 spid1 Recovering Database 'master' 97/05/19 20:03:21.21 spid1 Recovery dbid 1 ckpt (8460,2) oldest tran=(8460,1) 97/05/19 20:03:21.25 spid1 1 transactions rolled forward 97/05/19 20:03:21.37 spid1 Activating disk 'MSDBData' 97/05/19 20:03:21.37 kernel initializing virtual device 127, D:\MSSQL65\DATA\MSDB.DAT 97/05/19 20:03:21.37 spid1 Activating disk 'MSDBLog' 97/05/19 20:03:21.37 kernel initializing virtual device 126, D:\MSSQL65\DATA\MSDBLOG.DAT 97/05/19 20:03:21.37 spid1 Activating disk 'RMACLAIMS' 97/05/19 20:03:21.37 kernel initializing virtual device 2, D:\RMACLAIMS.DAT 97/05/19 20:03:21.37 spid1 Activating disk 'RMACLAIMSLOG' 97/05/19 20:03:21.37 kernel initializing virtual device 1, D:\MSSQL65\DATA\RMACLAIMSLOG.DAT 97/05/19 20:03:21.37 spid1 Activating disk 'TEMPdB' 97/05/19 20:03:21.37 kernel initializing virtual device 4, D:\TEMPdB.DAT 97/05/19 20:03:21.38 spid1 server name is 'RMASQL' 97/05/19 20:03:21.41 spid1 Recovering database 'model' 97/05/19 20:03:21.41 spid1 Recovery dbid 3 ckpt (338,19) oldest tran=(338,0) 97/05/19 20:03:21.53 spid1 Clearing temp db 97/05/19 20:04:37.04 kernel Read Ahead Manager started. 97/05/19 20:04:37.05 kernel Using 'SQLEVN60.DLL' version '6.00.000'. 97/05/19 20:04:37.14 kernel Using 'OPENDS60.DLL' version '6.00.01.02'. 97/05/19 20:04:37.14 kernel Using 'NTWDBLIB.DLL' version '6.50.232'. 97/05/19 20:04:37.15 ods Using 'SSNMPN60.DLL' version '6.5.0.0' to listen on '\\.\pipe\sql\query'. 97/05/19 20:04:37.15 ods Using 'SSMSRP60.DLL' version '6.5.0.0' to listen on ''. 97/05/19 20:04:39.08 spid11 Recovering database 'pubs' 97/05/19 20:04:39.09 spid11 Recovery dbid 4 ckpt (788,12) oldest tran=(788,11) 97/05/19 20:04:39.12 spid12 Recovering database 'msdb' 97/05/19 20:04:39.13 spid13 Recovering database 'RMACLAIMS' 97/05/19 20:04:39.14 spid11 1 transactions rolled forward in dbid 4. 97/05/19 20:04:39.15 spid13 Recovery dbid 7 ckpt (4096003,16) oldest tran=(4096003,15) 97/05/19 20:04:39.16 spid12 Recovery dbid 5 ckpt (3591,28) oldest tran=(3591,27) 97/05/19 20:04:39.22 spid12 1 transactions rolled forward in dbid 5. 97/05/19 20:04:39.23 spid13 1 transactions rolled forward in dbid 7. 97/05/19 20:04:39.49 spid1 Recovery complete. 97/05/19 20:04:39.50 spid1 SQL Server's default sort order is: 97/05/19 20:04:39.50 spid1 'nocase' (ID = 52) 97/05/19 20:04:39.50 spid1 on top of default character set: 97/05/19 20:04:39.50 spid1 'iso_1' (ID = 1) 97/05/19 20:04:39.63 spid1 Launched startup procedure 'sp_sqlregister' 97/05/20 08:20:12.29 ods Error : 17824, Severity: 10, State: 0 97/05/20 08:20:12.29 ods Unable to write to ListenOn connection '\\.\pipe\sql\query', loginname 'RMANET_SHIRLEE', hostname 'SHIRLEY'. 97/05/20 08:20:12.25 ods OS Error : 232, The pipe is being closed. 97/05/20 08:20:12.36 spid16 Error : 1608, Severity: 21, State: 2 97/05/20 08:20:12.36 spid16 A network error was encountered while sending results to the front end. Check the SQL Server errorlog for more information. 97/05/20 08:21:09.67 spid13 DBCC TRACEON 3605, SPID 13 97/05/20 08:21:09.68 spid13 Checking space allocation for database 7 97/05/20 09:44:46.88 spid13 DBCC TRACEOFF 3605, SPID 13 97/05/20 10:52:53.64 ods Error : 17824, Severity: 10, State: 0 97/05/20 10:52:53.64 ods Unable to write to ListenOn connection '\\.\pipe\sql\query', loginname 'RMANET_RIBREND', hostname 'RICHARD'. 97/05/20 10:52:53.64 ods OS Error : 109, The pipe has been ended. 97/05/20 11:35:12.28 spid49 EXCEPTION_ACCESS_VIOLATION raised, attempting to create symptom dump 97/05/20 11:35:12.28 spid49 Initializing symptom dump and stack dump facilities 97/05/20 11:35:17.24 spid49 Unable to create symptom dump file due to error (0) 97/05/20 11:35:17.27 spid49 Error : 0, Severity: 19, State: 0 97/05/20 11:35:17.27 spid49 SQLDumpExceptionHandler: Process 49 generated access violation; SQL Server is terminating this process Herewith also the details of the hardware config: Systemboard Configuration ========================= Processor: Digital Alpha 21164, Rev. 4.0 (1 Processor) Cache: 2MB Memory: 256MB AlphaBios: 5.28 970120.1242 Hard Disk Config ================ Disk 0 NCR810 #0, SCSI ID 0 2007 MB DEC RZ28D (C) DEC0008 Partition 1 2001 MB NTFS Partition 2 6 MB FAT Disk 0 SWXR0 #0, SCSI ID 8 16364 MB MYLEX DAC960 Partition 1 16364 MB NTFS PCI Configuration ================= Device Name Device Type Revision Physical Slot ------------ ----------- -------- ------------- Intel 82375 PCEB EISA Bridge 5 PCI0-Embedded S3 Trio32/64 VGA 0 PCI0-2 NCRC810 SCSI 2 PCI0-3 Digital 21140 Ethernet 18 PCI0-5 NCRC810 SCSI 2 PCI1-Embedded Digital 21141 Ethernet 17 PCI1-2 Mylex DAC960 Mass storage 2 PCI1-3 NCRC810 SCSI 2 PCI1-4 EISA Configuration ================== Device Name Device Type Physical Slot ------------ ----------- ------------- DEC6400 Other Embedded SCSI Configuration ================== NCRC810 #0, SCSI ID 7, SCSI Bus 0 SCSI ID DEVICE SIZE DESCRIPTION ------- ------ ---- ----------- 0 Disk 0 2007MB DEC RZ28D DEC0008 NCRC810 #1, SCSI ID 7, SCSI Bus 1 SCSI ID DEVICE SIZE DESCRIPTION ------- ------ ---- ----------- 5 CD-ROM DEC RRD45 DEC 0436 NCRC810 #2, SCSI ID 7, SCSI Bus 2 SCSI ID DEVICE SIZE DESCRIPTION ------- ------ ---- ----------- 4 TAPE DEC TZ88 DECCC34 SWXCR #0, SCSI ID 7, SCSI Bus 3 SCSI ID DEVICE SIZE DESCRIPTION ------- ------ ---- ----------- 8 Disk 1 16364MB MYLEX DAC960 MC Bus Configuration ==================== McBus Module Name Module Type Revision Physical Slot ----- ----------- ----------- -------- ------------- 1 Memory 0 0000 MEM 2 CPU (Cached) 2 0004 CPU0 4 MCBus-PCI Bridge 6008 0232 ID0/1 5 MCBus-PCI Bridge 6000 0032 ID0/1 RAID 5 DISKS ============ RZ29B-VW REV 0016 RAID CONTROLLER FW 2.36 Any help appreciated. I also have a copy of a performance log and event log available at the time of the crash which I can make available. Regards Nico Pretorius MCS Support (South Africa)
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 377.1 | CSC32::HOEPNER | A closed mouth gathers no feet | Tue May 27 1997 15:01 | 197 | |
Does the mssqlserver service stop? Or do some processes die?
Attached are two articles referencing errors in the log:
PSS ID Number: Q165158
Article last modified on 04-07-1997
PSS database name: SQLSERVER
6.5 6.0 4.21a
WINDOWS
======================================================================
---------------------------------------------------------------------------
The information in this article applies to:
- Microsoft SQL Server, versions 4.21a, 6.0, and 6.5
---------------------------------------------------------------------------
SUMMARY
=======
In some circumstances, the following communication errors are logged in the
SQL server error log:
- Error : 17824, Severity: 10, State: 0
Unable to write to ListenOn connection '<servername>', loginname '<login
ID>', hostname '<hostname>'.
OS Error : 64, The specified network name is no longer available.
Error 1608 A network error was encountered while sending results to the
front end
Other relevant errors may follow, depending on the network library used for
the connection. In case of Named Pipes connections, errors "232 The pipe is
being closed" or "109 The pipe has been ended" may be logged. In case of
sockets based connections (TCP/IP or IPX/SPX), errors "10054 Connection
reset by peer" or "10053 Software caused connection abort" may be logged.
These errors indicate that the connection between the client and SQL Server
was broken for some reason while SQL Server was writing data to the front
end. Depending on the circumstances, a certain number of these errors may
be normal, and eliminating them is not always necessary. If you do not
receive any end user complaints or experience general connection problems,
you can ignore these errors.
MORE INFORMATION
================
Error 17824 "Unable to write to ListenOn connection" indicates that
problems have occurred while SQL Server attempted to write to a client
connection and failed. These communication problems may be caused by
network problems, or if the client computer has stopped responding or been
restarted. For more information regarding other communication errors,
please refer to the following article in the Microsoft Knowledge Base:
ARTICLE-ID: Q109787
Title : INF: SQL Communication Errors 17832, 17824, 1608, 232,
and 109
Error 17824 does not always indicate a network problem. The following are
the most common situations under which the error 17824 is generated, along
with the corresponding troubleshooting procedures.
- This error may occur if the users are restarting their client computers
if the application seems have stopped responding, so make sure they
don't do that. It may be that the server is taking a longer time to
process a long query. Once the client workstation is restarted, the
connections are broken ungracefully. Later SQL Server tries to respond
to the connection that has been dropped, and logs the message 17824.
- The network may be unstable; make sure it is stable. You can check this
by attempting to copy large files between the computer running Windows
NT Server and the client computer. If this test fails, then you are
running into problems with the physical network. Because the above
errors indicate a potential network issue, it is recommended that you
update the server, both Windows NT Server and SQL Server, to the latest
service packs. Please check the knowledge base articles for more
information on how to obtain the latest service packs for the Windows NT
Server operating system and SQL Server. It is also recommended that you
update the client components such as DB-Library, the ODBC driver, and
network library to the latest DLLs. In case of client computers running
16-bit Windows 3.1 or Windows for Workgroups on a Novell network, it is
necessary to obtain and install the latest MS-DOS and Windows drivers
from Novell.
- Error 17824 "Unable to write to ListenOn connection" may be a
consequence of other errors that caused the connection to drop. Check
the error logs for other errors within the same time frame as the 17824
error. If you find other errors, refer to SQL Server Books Online and
the Microsoft Knowledge Base for more information on these errors.
- Use sp_configure or the SQL Server Enterprise Manager to check the
Priority Boost and the SMP Concurrency configuration settings. Make sure
that these two configuration options are set to the default settings,
because deviating from the default settings may cause error 17824 under
some conditions.
For more information on why these settings may generate error 17824,
please refer to the following article in the Microsoft Knowledge Base:
ARTICLE-ID: Q111405
Title : INF: SQL Server and Windows NT Thread Scheduling.
- The error 17824 may be generated due to application problems. One major
cause is running into a lock or block situation. In this case, a process
holds a lock on a page or a table, and that lock is not released right
away, due to an uncommitted transaction or a long query. This may cause
all other processes requesting the same table to be blocked, and the
client application to seem to stop responding. If the user then either
uses 'End Task' to close the application or restarts the workstation,
you may receive error 17824 on the server.
To find out if the application is causing a lock or block problem on the
server, use the sp_who and sp_lock stored procedures when the client
computers seem to stop responding or when the error 17824 starts to
appear in the SQL Server error logs. If the client workstation has
stopped responding, open a command-line ISQL connection on the server
itself using the local pipe, and use these stored procedures to check
for a blocking situation.
For more information on detecting and resolving blocking problems, please
refer to the following article in the Microsoft Knowledge Base:
ARTICLE-ID: Q162361
Title : INF: Understanding and Resolving SQL Server Blocking
Problems
If you have checked all of the above points and the problem still persists,
contact SQL Server support for further troubleshooting assistance.
Additional query words: Internet IIS WAN symmetric multiprocessor SEM
Advanced options
======================================================================
Keywords : kbtshoot kbusage SSrvGen
Version : 6.5 6.0 4.21a
Platform : WINDOWS
Issue type : kbhowto
Resolution Type : Info_Provided
=============================================================================
Copyright Microsoft Corporation 1997.
| |||||