on 12-04-2008 1:52 PM
Hi all,
I'm a new comer in MaxDb LiveCashe configuration.
We have implemented recently an SCM (4 landscape°) .... everything works fine ... But since 3 days our LiveCashe production hangs
Connexion is possible anymore (via lca03, Database Manager and at the OS level)
Maybe it is a full log situation , I have no error message but the probleme is that I can't solve the issue as the connexion is not possible.
Thank you for your help ( it's really critical now :-(( )
Cheers
Yacine
Hello Yacine,
1) You could start the analysis of the issue by yourself, login to the liveCache server at OS
level:
a) Check errors in Xserver, dbmsrv protocols;
b) check errors in knldiag or KnlMsg files
c) Run 'x_cons <SID> show active' & check the active tasks in liveCache
d) check if you could connect to the liveCache using dbmcli tool on the liveCache server
2)What is your OSS ticket number?
3)"But the SAP support is not so good that it was."
If the case was critical, did you create the ticket with very high priority?
Did you open the connection to your system via OSS?
Did you provide all information need for the analysis of the reported issue?
Thank you & best regards, Natalia Khlopina
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi Natalia
1) You could start the analysis of the issue by yourself, login to the liveCache server at OS
level:
a) Check errors in Xserver, dbmsrv protocols;
As I told, I know how to deal with Oracl but for me MaxDB is a new technology
So I don't know where to find Xser, dbmsrv protocol
b) check errors in knldiag or KnlMsg files
In knlMsg I have no error as no startup is possible with the command dbmcli
This command hangs
I can wait 2 3 hours or more I have no result
When I abort the command I have the following output
eowyn:l90adm 21> dbmcli -d L90 u control,gl0bulus
Error! Connection failed to node (local) for database L90:
Reading header from pipe:(32:Broken pipe
eowyn:l90adm 22>
So Live Cashe can not be started
Information important : We are in SGeSAP configuration and binary,data are stored in a volume group whih is mounted via NFS;
This solution has worked fine until the21th of november.
c) Run 'x_cons <SID> show active' & check the active tasks in liveCache
eowyn:l90adm 1> x_cons L90 show active
SERVERDB: L90
ID UKT UNIX TASK APPL Current Timeout Region Wait
tid type pid state priority cnt try item
Post Mortem Analysis for ServerDB L90 using kernel ***
Console command finished (2008-12-04 21:38:40).
d) check if you could connect to the liveCache using dbmcli tool on the liveCache server
Impossible to connect
For instance when i perform the following command works
eowyn:l90adm 8> dbmcli -ux SAPR3,SAP -ul
OK
DEFAULT SAPR3
1L90l90lcpkg CONTROL/GL0BULUS
ag L90ADM/GL0BULUS
c CONTROL
But this one hangs
eowyn:l90adm 9> dbmcli -U c
2)What is your OSS ticket number?
I have 2 tickets 0000987768 and 0001008659
Best regards
Yacine
Hello,
As I saw that the delayed < ~ 10days > with update in ticket 987768, therefore SAP support was not able to help you.
You created 1008659 ticket with medium priority. Is it the same issue as in ticket 987768?
If yes, then it's not good idea to create the new ticket, please update the ticket 987768 with requested information by SAP support .
Thank you and best regards, Natalia Khlopina
Hi Natalia
No it's not the same ticket
The first tickect is related to issues reported when this instance was OK
Indeed when the host of the package (instead of the host name) was set via LC10 ,then the connexion could not be possible.
Now the problem is different , nothing works
No connexion is possible and startup of the DB is not possible
Thanks for your answer
Yacine
Hi all,
Helped me to solve this issue.
The problem was caused by files created during a previous startup and after a crash some files were remaining.
For instance :
[root@eowyn /]# ll /sapdb/data/fifo/
total 0
prw-rw---- 1 sdb sdba 0 Nov 20 09:18 xserver_eowyn_logger
prw-rw---- 1 sdb sdba 0 Nov 20 09:18 xserver_eowyn_update
prw-rw---- 1 sdb sdba 0 Nov 11 22:54 xserver_faramir_logger
prw-rw---- 1 sdb sdba 0 Nov 11 22:54 xserver_faramir_update
I have deleted them
[root@eowyn fifo]# rm xserver_*
rm: remove fifo `xserver_eowyn_logger'? y
rm: remove fifo `xserver_eowyn_update'? y
rm: remove fifo `xserver_faramir_logger'? y
rm: remove fifo `xserver_faramir_update'? y
Then these filesrelated to shared memory were also moved
mv /sapdb/data/wrk/L90.dbm.shi /sapdb/data/wrk/L90.dbm.shi.old
mv /sapdb/data/wrk/L90.dbm.shm /sapdb/data/wrk/L90.dbm.shm.old
And then the connexion was possible.
Thanks to Natalia for her help and quick feedback
Yacine
Solution :
Delete the old remaining files and shared memory
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
HI Yacine,
> Thank you for your help ( it's really critical now :-(( )
whenever it's really critical the forum is the wrong place to go to.
Open a support message with priority HIGH or VERY HIGH (depending on the criticality) and have the support to take a look into this.
regards,
Lars
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
83 | |
24 | |
12 | |
9 | |
7 | |
6 | |
5 | |
5 | |
5 | |
4 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.