errpt, abap dump

Basis (Basis Technology Modules: Basis Component/System Administration, GUIs)

Moderators: Snowy, thx4allthefish

Post Reply
yls177
Posts: 4016
Joined: Thu Oct 24, 2002 9:12 pm

errpt, abap dump

Post by yls177 » Thu Mar 27, 2003 10:13 pm

in my prod server, i have errors in errpt with the program as disp+work frequently... i think it is related to the abap short dumps... with st22, there are existing abap dumps. but besides listing these abap dumps, any further steps that i could undertake? thanks

Snowy
Posts: 28790
Joined: Mon Oct 21, 2002 2:33 pm
Location: 3.1415926535

Post by Snowy » Mon Mar 31, 2003 10:07 am

what is the meaning of "errpt" ?

Snowy

yls177
Posts: 4016
Joined: Thu Oct 24, 2002 9:12 pm

Post by yls177 » Mon Mar 31, 2003 11:08 am

snowy, from your posts, i have observed that u are on sun platform. errpt is the same as dmesg.... u will get what i mean?

Snowy
Posts: 28790
Joined: Mon Oct 21, 2002 2:33 pm
Location: 3.1415926535

Post by Snowy » Mon Mar 31, 2003 11:24 am

Are you getting "red" numberse in SM50? When WP are dying, error count inceases by 1 and is shown in red. If this is the case, investigate the appropriate dev_w* log file for more info.

Snowy

yls177
Posts: 4016
Joined: Thu Oct 24, 2002 9:12 pm

Post by yls177 » Tue May 13, 2003 4:34 am

errpt entries appear on may 13 16:28. from dev_w0, before and after...messages are as belows.

before may 13 16:28

D Tue May 13 16:05:43 2003
D *** ERROR => CuaSubstitute: empty variable name. [dycuatab.c 2334]
D *** ERROR => CuaSubstitute : fcode/textno -> SP§§ /001 [dycuatab.c 2338]
D *** ERROR => CuaSubstitute : modulpool -> SAPLMGMM [dycuatab.c 2341]
M
M Tue May 13 16:16:24 2003
M *** WARNING => ThINoGetInterval: no entry found for cli/obj/subobj/range = 040/SD_PICKING/ /01
M
M Tue May 13 16:25:11 2003
M *** WARNING => ThINoGetInterval: no entry found for cli/obj/subobj/range = 040/SD_PICKING/ /01


after may 13 16:28

M Tue May 13 16:25:11 2003
M *** WARNING => ThINoGetInterval: no entry found for cli/obj/subobj/range = 040/SD_PICKING/ /01
M
M Tue May 13 16:29:48 2003
M *** WARNING => ThINoGetInterval: no entry found for cli/obj/subobj/range = 040/SD_PICKING/ /01
M
M Tue May 13 16:30:21 2003
M *** WARNING => ThINoGetInterval: no entry found for cli/obj/subobj/range = 040/SD_PICKING/ /01


from errpt, i got
SIGNAL NUMBER
4

and from note 335562, it says
"ora-7445 [argument1] [argument2]
argument1:
generally gives the kill level of the process. This means the kill level can be read on most Unix derivatives by looking it up in file /usr/include/sys/signal.h
SIGHUP 1 /* hangup */
SIGQUIT 3 /* quit (ASCII FS) */
SIGILL 4 /* illegal instruction (not reset when caught)
and ..................."

but in my alert file there is no ora-7445

Snowy
Posts: 28790
Joined: Mon Oct 21, 2002 2:33 pm
Location: 3.1415926535

Post by Snowy » Tue May 13, 2003 7:48 am

3 things come into mind:

1. upgrade kernel
2. check if you have bad memory on server
3. check latest OS patches

Snowy

thx4allthefish
Posts: 5694
Joined: Sat Oct 26, 2002 6:18 pm
Location: barolo barrel

Post by thx4allthefish » Tue May 13, 2003 8:30 am

i would like to hear what those abap dumps say. but by looking at the error-messages you pasted i would say you have some problem with the number ranges/number buffering especially when the taskhandler is trying to get a new number from an interval (ThINoGetInterval) (concerning material master (SAPLMGMM). Would you mind checking that?

yls177
Posts: 4016
Joined: Thu Oct 24, 2002 9:12 pm

Post by yls177 » Tue May 13, 2003 9:38 am

Snowy wrote:3 things come into mind:

1. upgrade kernel
2. check if you have bad memory on server
3. check latest OS patches

Snowy
i have no problems with 1 and 2. but how do i check whether i have bad memory on server.
thx4allthefish wrote: i would like to hear what those abap dumps say. but by looking at the error-messages you pasted i would say you have some problem with the number ranges/number buffering especially when the taskhandler is trying to get a new number from an interval (ThINoGetInterval) (concerning material master (SAPLMGMM). Would you mind checking that?
hmmm, i will try my very best to look at the abap dumps.. and try to post some useful messages.....

also, how do i check the process of getting a new number from an interval (ThINoGetInterval) (concerning material master (SAPLMGMM).

thanks

Snowy
Posts: 28790
Joined: Mon Oct 21, 2002 2:33 pm
Location: 3.1415926535

Post by Snowy » Tue May 13, 2003 10:04 am

most OS have diagnostics program that can be run. Check with IBM.

good luck.

Snowy

thx4allthefish
Posts: 5694
Joined: Sat Oct 26, 2002 6:18 pm
Location: barolo barrel

Post by thx4allthefish » Tue May 13, 2003 10:05 am

use transaction snro - object sd_picking - is buffering activated? how many entries? then go to transaction sdpi - check on interval 01- are there enough free numbers left? do u use internal or external numbering?
check in transaction st02 -> detailed analysis menu -> other buffers -> number ranges ...

cli/obj/subobj/range = client = 040 -> object = sd_picking -> subobject = blank -> range = 01.

Post Reply