cancel
Showing results for 
Search instead for 
Did you mean: 

Error 22 for write/read access to a file - in system log SM21

Former Member
0 Kudos

Hi

We have a 640 (SP19) R/3 development system that keeps crashing. Upon restart we can see that the last entries in the system log (SM21) are as follows:

Error 22 for write/read access to a file. File =

File is often blank, sometimes it has a username with 010 appended like so:

Error 22 for write/read access to a file. File = BIDDAPPA_D010

Details of message are:

When it accessed a file, the TemSe software received an unexpected

error message from the operating system (or file system). The message

contains the current file name and the value of the "errno" variables

of the C system.

-checked TemSe via SP12 and there were lots of spool inconsistencies. These have now been cleared.

However, this morning we still have lots of the same errors in SM21 (although the system hasn't fallen over.... yet...).

Nothing in ST22 - no related dumps - can't see anything of use in the worklog trace files.

Any ideas?

Regards

Ross

Accepted Solutions (1)

Accepted Solutions (1)

nelis
Active Contributor

Error 22 for write/read access to a file. File = BIDDAPPA_D010

As your <SID>adm user make sure you can access /sapmnt/<SID>/global directory. If this directory is a network mount and you are storing data in temse on the file system(rspo/store_location = G) you could have the issues you mentioned when there is either a network or permissions problem. My guess is permissions.

--

Nelis

Former Member
0 Kudos

Thanks Nelis... but rspo/store_location is set to db.

Checked the filesystem anyway - no permissions issues.

JPReyes
Active Contributor
0 Kudos

For the errors above check Note 385894 - Message 00 560: Logon of user & failed

But I don't think they are related to the main issue here,

Regards

Juan

Answers (2)

Answers (2)

Former Member
0 Kudos

Not really sure what happened here... seemed to fix itself... or the developers did something..??!

JPReyes
Active Contributor
0 Kudos

have you checked the developer traces for more info?.... I have a feeling that is related to some trace put on a user/object that hasn't been disabled maybe.

Regards

Juan

Former Member
0 Kudos

Hi Juan

By the "developer traces" do you mean dev_w25 etc in /usr/sap/SID/DVEBMGS00/work ? I have already checked those (what I meant by "worklog trace files"). Or do you mean something else?

There were errors in dev_w25 but they were meaningless and searches on the errors didn't yeild any useful information.

Regards

Ross

JPReyes
Active Contributor
0 Kudos

Yes, I meant dev_*... you might not be lloking into the right trace...

Regards

Juan

Former Member
0 Kudos

Looking at system log the latest errors occurred in background work process 28 but looking at dev_w28 I don't see any errors... however, looking at the latest file dev_w25 I can see these errors at the time of the problems:

L Tue Dec 8 13:16:28 2009

L *** ERROR => BtcAuthSet: dy_signi-call failed, rc: 2 [btcjcntl.c 1025]

L

L Tue Dec 8 13:17:28 2009

L *** ERROR => BtcAuthSet: dy_signi-call failed, rc: 2 [btcjcntl.c 1025]

L

L Tue Dec 8 13:18:23 2009

L *** ERROR => BtcAuthSet: dy_signi-call failed, rc: 1 [btcjcntl.c 1025]

L

L Tue Dec 8 13:18:28 2009

L *** ERROR => BtcAuthSet: dy_signi-call failed, rc: 2 [btcjcntl.c 1025]

L

L Tue Dec 8 13:18:43 2009

L *** ERROR => BtcAuthSet: dy_signi-call failed, rc: 1 [btcjcntl.c 1025]

Not much useful information...

Can see in another trace:

A Tue Dec 8 13:14:29 2009

A WP has reached abap/heaplimit = 40000000 bytes

- now that is worth looking into... but not sure it's related....