[Info-vax] Need help tracking down a MUTEX state

[Jim]

On Aug 19, 12:47 pm, "Ade" <adrian.birk... at blueyonder.co.uk> wrote:
> Hi,
>
> We have a VAX 6000-610 running VMS 6.1. On occasion the system hangs so I've
> had the operators crash the system as per instructions found elsewhere.
>
> The dump of the system seems fine except for one process in MUTEX state
> prio=16, mutex count=0 and PCB$L_EFWM=JIB of this process. On the quota
> front the only 'anomaly' is that the BUFIO byte count/limit is at 192/65216.
> None of the other usual suspect quotas are diminished.
>
> A show proc/chan shows a busy channel to the system disk but no associated
> file id.
>
> Can someone please post a document or pointers so as to how to trace what's
> actually going on?
>
> Thanks,
>
> Ade

Looks like you've already found the problem - looks like those 192
remaining bytes of bytlm aren't enough to perform the next IO. You'll
need to increase BYTLM. When the EFWM value matches the JIB address
then you’ve got a quota problem. Then, you need to determine which one
– the JIB flags (jib$l_flags) will tell you. A value of 1 indicates
bytlm, a 2 indicates tqelm, and a 3 indicates both have been
exhausted.

$ analyze/system
SDA> set proc/id=xxxxxx	! whatever the target PID is…
SDA> sh proc
SDA> ! see if the EFWM = JIB address
SDA> ! if it does then
SDA> read sys$loadable_images:sysdef.stb
SDA> format/type=jib JIB	! where JIB is the process JIB address
SDA> ! look for the JIB$L_FLAGS and note the associated value