Tried to
reorg rebuild
the whole table but not enough space in the database to do that
The only way I could fix it was to copy out the data, delete the data and re-insert it.
A truncate on the partition was incredably slow - I killed it before it completed.
I thought this would be quick its only 5000 pages - but having to update the global index means
the time for a truncate of a partition isn't much different from a delete.
(Obviously speed of page read is the most expensive operation)
Hello Maria Victoria,
Your are right, I'm not using ASE for SAP applications and thanks because you moved the discussion to the SAP ASE for custom applications but I'm sure I posted the discussion on " SAP ASE for custom applications".
By the way, the batch doesn't execute bulk insert. They are 14 processes running in parallel, modifying data.
The strange is after this issue, I have just one way to clear the transaction log. I must execute "dump tran base with truncate_only" and ASE begin to show different kinds of error messages and I need to kill the dataserver by OS (shutdown with no wait, doesn't work).
Thanks,
Juan Carlos.
One thing you might want to check.....since this is an upgraded server, you might need to increase the stack and stack guard size. I have seen issues with procs not executing (e.g. sp_sysmon) on upgraded ASE's that had smaller stacks previously:
exec sp_configure 'stack size', 194560
exec sp_configure 'stack guard size', 20480
...that may be enough to get sp_sysmon and MDA working to allow better diagnostics
You also said that there wasn't any long running open txns. As you are aware, of course, checkpoint can't truncate passed the truncation point (oldest open txn)....it might be interesting to see where things are with respect to that....you might want to look up loginfo() and during one of the batches do something like:
declare @n int
set @n=0
while @n<=1000
begin
select loginfo(<dbname>,first_page) as first_page,
loginfo(<dbname>,checkpoint_page) as checkpoint_page,
loginfo(<dbname>,oldest_active_transaction_page) as oldest_txn_page
waitfor delay '00:00:01'
select @n=@n+1
end
go
.....by letting it run for 1000 secs, there should be several checkpoints fire....
I also wonder what happens if you do:
use master
go
sp_dboption dbname, 'trunc', false
go
use dbname
go
checkpoint
go
use master
go
sp_dboption dbname, 'trunc', true
go
use dbname
go
checkpoint
go
...in otherwords, I wonder if the bit is set in the system catalog, but the dbtable structure on disk is messed up. You might be able to verify this by checking with dbtable for DBT_AUTOTRUNC:
1> use demo_db
2> go
1> dbcc traceon(3604)
2> go
DBCC execution completed. If DBCC printed error messages, contact a user with System Administrator (SA) role.
1> dbcc dbtable('demo_db')
2> go
DBTABLES (ACTIVE):
DB Table at: 0x0000000029C4C8C0
dbt_dbid=8 dbt_stat=0xc (0x0008 (DBT_AUTOTRUNC), 0x0004 (DBT_SELBULK))
dbt_extstat=0x0 (0x0000)
dbt_stat2=0xffff8003 (0x8000 (DBT2_MIXED_LOG_DATA), 0x0002 (DBT2_NOACCOUNT), 0x0001 (DBT2_LOGFULLABT))
dbt_stat3=0x20800 (0x00020000 (DBT3_SYSPARTITIONS_EXISTS), 0x00000800 (DBT3_DELAYED_COMMIT))
dbt_stat4=0x0 (0x00000000)
dbt_runstat=0x0(0x0000)
dbt_state=0x2(0x0002 (DBST_ACTIVE)) dbt_keep=1 dbt_hdeskeep=0 dbt_next=0x0000000029C44C20 dbt_systask_keep=0 dbt_detachxact_keep 0
dbt_lock=0 dbt_dbaid=1
dbt_verstimestamp= Oct 22 2009 6:28AM dbt_dbname=demo_db
dbt_logrows=0
dbt_lastlogbp=0x00000000438A06F0
dbt_logsema=0000000003BAEE40
dbt_nextseq=12 dbt_oldseq=12
....if not in dbtable, then doing the dboption toggle might help. Otherwise, something is affecting the checkpoint spid in that it isn't running at all - e.g. someone changed the user priority or something.....
If it is the views, the most common situation is an upgraded ASE- once the view is dropped/recreated in the new version, it functions fine.....
Thanks
How do you tell if its the views ?
monCachedProcedures is only showing 2 ObjectTypes on all our servers
ObjectType Count MemUsageKB
-------------------------------- ----------- ---------------------
trigger procedure 66 13282
stored procedure 13404 2782438
Hi,
Were you able to reproduce this ?
I'll log this as an incident.
Thanks
Mike
Hi Mike,
correction request 782871 does not fix the full scope of the problem, so we now also have 786594 to address this.
Let me know the incident number will associate this with the 786594.
Niclas
Thanks - I've raised one here.
606256 / 2015
Hi
Last night user was locked out of ASE server (15.5 ESD 2) briely -- Application Had following error:
DBError: [20011]: Maximum number of DBPROCESSes already allocated
Around the time the error was reported I noticed strange entry in ASE error log -- for about 1 minute the hour and minute entries were not written to log:
00:14:00000:06133:2015/07/16 20:52:40.83 kernel Cannot read, host process disconnected: BT4-TRD-STC 00000123 spid: 6133
00:02:00000:06524:2015/07/16 20:52:40.84 kernel Cannot read, host process disconnected: BT4-TRD-STC 00000123 spid: 6524
00:06:00000:09674:2015/07/16 :11.90 kernel Cannot read, host process disconnected: APPW54-STC 00000123 spid: 9674
00:16:00000:07510:2015/07/16 :27.69 kernel Cannot read, host process disconnected: APPW54-STC 00000123 spid: 7510
00:04:00000:01925:2015/07/16 :27.69 kernel Cannot read, host process disconnected: APPW54-STC 00000123 spid: 1925
00:00:00000:00283:2015/07/16 :43.87 kernel Cannot read, host process disconnected: appw54-stc spid: 283
00:09:00000:05736:2015/07/16 :43.88 kernel Cannot read, host process disconnected: appw54-stc spid: 5736
00:20:00000:06548:2015/07/16 :43.92 kernel Cannot read, host process disconnected: appw54-stc spid: 6548
I have not seen that in log before ..
at time this happened there were number of maintenance actions running,..
any thoughts..?
Thanks
Jim
Documented error, it's in the troubleshooting guide
The Backup Server runserver file, the interfaces file, and the sysservers table in the master database do not have a consistent name for the Backup Server.
or, more likely if it's an intermittent error
You have hit a network limit for the Backup Server process. Typically, this limit is set to 25.
Start Backup Server with the -Nnn flag to change the network limit. For example, -N30 increases the limit to 30.
Although this is an error reported by the Backup Server, it is not unique to it. The error can be raised by any client built using Open Client DB-Library. DB-Library has a default of 25 DBPROCESS structures, but can be configured for more. It sounds as if your client exceeded it's current setting.
Details on this setting can be found in the
Software Developer's Kit 15.7 > Open Client DB-Library/C Reference Manual > Routines
http://infocenter.sybase.com/help/topic/com.sybase.infocenter.dc32600.1570/html/dblib/X42139.htm
Errors
The
dbopen call will return NULL if any of the following errors occur. These errors
can be trapped in the applications error handler (installed with
dberrhandle.)
Error Condition/cause
-------
-----------------------------------------------------------------------
SYBEDBPS
Maximum number of DBPROCESSes already allocated.
Note that an application
can set or retrieve the maximum number of DBPROCESS structures with
dbsetmaxprocs and
dbgetmaxprocs
dbsetmaxprocs
http://infocenter.sybase.com/help/topic/com.sybase.infocenter.dc32600.1570/html/dblib/X47066.htm
Description:
Set the maximum number of simultaneously open DBPROCESS structures.
Syntax:
RETCODE dbsetmaxprocs(maxprocs)
int
maxprocs;
Parameters
maxprocs
The new limit on simultaneously open
DBPROCESS structures for this particular program.
Returns
SUCCEED or
FAIL.
Usage
A DB-Library program has a maximum number of
simultaneously open DBPROCESS structures. By default, this number is 25. The
program may change this limit by calling dbsetmaxprocs.
The program may
find out what the current limit is by calling dbgetmaxprocs.
The missing hour and minute fields in the errorlog are certainly strange, I have not seen that before. It may or may not be related to the DBPROCESS error.
When you say there were a number of maintenance actions running, do you mean someone was running DBCCs, REORGS, or DUMPs, or was it that sp_who or sysprocesses showed some number of connections as "MAINTENANCE TOKEN"? The later is often seen when a connection is early in the login process as it is the 0th element in an array of possible values and so shows up when the field is uninitialized. In that case, I think it may be related as the client may be initiating a login but then hitting the DBPROCESS error and breaking the connection early in the process.
-bret
They way I read this, Edge replaces SBE and SBE will no longer be available (in the future).
Edge lets you use only 4 cores instead of 8 for SBE (Edge boasts unlimited "CPU chip" count, but that's useless/irrelevant if only 4 cores total are allowed per host)
And although Edge allows unlimited database size, in practice, larger databases need will need to use a few cores for running backups. So if you need 2 stripes to get good dump throughput, that will use up 2 cores, and the dataserver will be effectively limited to remaining 2 cores (ie., 4 cores minus the number of dump stripes)
So Edge is really for smaller applications than SBE. But there all the extra security, encryption, and partitioning features available in Edge.
Hi Jeff,
Thanks for your answer.
I am going to run the batch with your recommendations.
I'll give your feedback about it.
Juan Carlos.
Hello Jeroen,
No, nothing is reported in the errorlog.
I have altered that message. Executing the query still doesn't report something in the errorlog.
We have this with only one query, it's a small table, so I copied it into a table with a different name, renamed the original table and created a view on the new table with the original name.
That worked and eliminated our production problem.
The problem was still reproducable on the old table.
Difference with the new situation is that triggers were added to the new table and didn't cause problems.
So I dropped and recreated the update trigger on the old table, and then the problem disappeared.
Conclusion is that there must have been some corruption in the trigger code.
The original situation is restored and the query is running fine.
Thanks,
Luc.
Hi --
We are still seeing this occasionally. Happened earlier today -- server new connections wee locked for about 45 seconds --
saw following in log:
00:07:00000:04458:2015/07/20 11:21:12.10 server Update of syslogins failed
00:04:00000:03632:2015/07/20 11:21:12.10 server Update of syslogins failed
00:18:00000:07700:2015/07/20 11:21:12.11 server Update of syslogins failed
00:16:00000:04445:2015/07/20 11:21:12.11 server Update of syslogins failed
and
00:05:00000:05915:2015/07/20 11:21:49.53 server Unable to read in login record.
00:19:00000:04791:2015/07/20 11:21:50.28 server Unable to read in login record.
We did implement some changes per older ticket
set configuration for enable last login updates' to 0
Set lastlogincount to 0 where was null ..
The threshold needs to generate a message (eg, "print @msg"). Since the threshold has no front-end to send the print message to, the output goes the errorlog.
I normally have the threshold print a few messages ... initial message stating db/segment/space_used/%used message about what it's going to do (eg, dump tran/truncate; dump tran to a file; etc), and a final message stating db/segment/space_used/%used (will need to recalculate after whatever operation you perform).
Hi, I am trying to troubleshoot a colleague's strange issue with java isql and ASE15.5. Periodically, any query or stored procedure (like sp_who) returns 0 rows. Simple select queries from system or user tables, there is no issue with the queries. I can run the same queries in DOS isql and results are returned. Connection to ASE is established, so I have no idea what's going on. Playing around with it, I switched the option "Automatically refetch results" off and on under tools->options and the queries start returning results. However, when I close and restart isql, the same issues happens and I have to switch that option off and on (or leave it off) again. I have no idea why.
Anyone have any ideas what's going on?
thanks,
Hi all,
I have been using ASE linux 15 from to connect from my application and recently i installed ASE linux 16 developer version and loaded all data from my old one to the new one and when i try to connect from the terminal (say putty) i am able to connect to the server and perform all the operations.
But when i try to connect from the application it is giving me error saying invalid user name.
But with the same user i am able to connect to the old db and also to the new DB through terminal.
I have configured the LDAP authentication and changed the interface files.
I added a new user using the below command..
"create login username with password"
is there anything that i am missing ?
Can anyone help me out why i am not able to login through my application to the new DB.
is it a problem with the LDAP authentication ?
Any reference on this would also be of great help.. 
thanks for the reply gaurav..i have checked the error log and it has no errors..
We can see lots wait event about "CMCC task waiting for...".
What is CMCC full represent? Maybe.....Cluster Memory Consistency Check?
Any Url talk talk about ASE CE CMCC?
Thanks