Work Process tries to restart but fails - Operating system call fork:restar

[Former Member]

Hi

We have a ECC 5.0 system running on v5r4 (6.40 kernel level 159). Some of the work processes fail and then try to restart and fail again. Thus there are many job log entries in both SAP and OS/400.

The OS/400 logs mention the PASE (unix executable environment) and MCH3203.

SAP mentions Operating system call fork:restart

OSS has note #884359 which mentions a palitative but despite being at 6.40 level 159 the parameter profile as/4/wp_max_restart is said to be "unknown" by RZ10.

We are making sure we are up to date with the APAR II14126 just in case this is the issue.

It did start after the time change but we are not sure this is the root cause.

We are also considering getting upto kernel 171.

Anyone else seen this issue?

thanks

Matthew Milne

[christian_bartels · Accepted solution]

We see those problems once a while, and usually they are gone when you restart the instance or after an IPL. You should report the problem to IBM, they will probably be interested in the LIC log entries that come along with the error (STRSST - Start a Service Tool - Licensed Internal Code Log). II14126 is probably a good start (for example, MF40759).

The profile parameter as4/wp_max_restart is only known to the kernel, because it is not intended as a permament setting, but only as a temporary workaround for the problem in the PASE environment. That's why RZ11 does not know about the parameter and reports it as unknown.

Kind regards,

Christian Bartels.