Operating System - OpenVMS
1748182 Members
3569 Online
108759 Solutions
New Discussion юеВ

Re: What is -SYSTEM-F-HWM_STALL?

 
SOLVED
Go to solution
Malcolm Wade
Valued Contributor

What is -SYSTEM-F-HWM_STALL?

I have a VMS NetBackup client partially failing every so often on a 4GB+ RDB backup (.RBF) file with:

%RMS-E-READERR, error reading file file-name
-SYSTEM-F-HWM_STALL, internal I/O stalled by highwater marking activity

file-name is the actual file spec.

What is HWM_STALL? HELP/MESSAGE doesn't know about it and a google search produces a piece of code from the source I think with Andy Goldsteins name against it.

Yes, Highwater marking is enabled on the volume on which the file is found.

Thanks,
Malcolm
8 REPLIES 8
Volker Halle
Honored Contributor

Re: What is -SYSTEM-F-HWM_STALL?

Malcolm,

SS$_HWM_STALL is signalled by the XQP (file system). As such, I would assume this to be an OpenVMS internal problem. Make sure you've installed the most recent F11X and XFC patches and are running a supported version of OpenVMS. If the problem still happens, log a call with HP.

As a workaround, consider turning off HWM.

The 'source' you're referring to is just including a list of possible OpenVMS system service failure codes (and the internal revision history).

Volker.
Willem Grooters
Honored Contributor

Re: What is -SYSTEM-F-HWM_STALL?

I'm not familiar with NetBackup, so just a guess where to look.

It could be that NetBackup writes information on that same disk, or at least, tries to do that, and that could cause this problem when the disk is heavy loaded, and , perhaps, very full. Also, it might be that the NetBackup process tries to access data not written by the owner of the process.
Turning that facility off could be a solution. Perhaps raining privileges could do the trick as well.

Willem
Willem Grooters
OpenVMS Developer & System Manager
Galen Tackett
Valued Contributor
Solution

Re: What is -SYSTEM-F-HWM_STALL?

Willem,

Thanks for the expression "raining privileges." I've seen so many applications that do this, but I've never had such a concise and clever name for it.
:-)

Galen
John Gillings
Honored Contributor

Re: What is -SYSTEM-F-HWM_STALL?

Malcolm,
Please log a call. This looks interesting.
A crucible of informative mistakes
Eberhard Wacker
Valued Contributor

Re: What is -SYSTEM-F-HWM_STALL?

Hi Malcolm,
as said above make up a call and
as said above turn off highwater marking on this or all volumes if your system is not a top security related one and you want to gain better performance (it can be made dynamically on the fly with set vol/nohigh).
Besides this:
is the RBF file in the same directory as the database itself ? And is there heavy database activity during the backup operation ?
Then create it on another directory (or better another disk with few activity during the backup period).
Cheers
EW
Malcolm Wade
Valued Contributor

Re: What is -SYSTEM-F-HWM_STALL?

Thanks to all responses.

To try and answer all questions; VMS V7.3-2 with Update v3 applied.

The RBF file is actually a copy of the pre-overnight Production database backup which is copied in via DECnet after it is created. Purely and simply there for testing purposes.

HWM is on; one of my applications needs it on otherwise it breaks! A similar post-overnight backup on another disk with HWM turned on never has any issues.

I'll log a call.
Volker Halle
Honored Contributor

Re: What is -SYSTEM-F-HWM_STALL?

Malcolm,

the XQP tries to restart pending writes, when a HWM erase operation on the file has completed. If the IO to be re-started is an 'internal IO', it returns SS$_HWM_STALL.

One of the possible 'internal QIOs' may come from the XFC. Could you try a SET FILE/NOCACHE on the .RBF file before running NetBackup ?

And are you sure, that noone is (still) writing to the .RBF file, when NetBackup may be reading it ?

With V7.3-2, you could use the IO$SDA extension to trace IOs. If you can catch the NetBackup error status immediately, you could stop IO tracing and should find the IO in the trace:

$ ANAL/SYS
SDA> IO LOAD
SDA> IO START TRACE

Once NetBackup exists with error, use:

$ ANAL/SYS
SDA> STOP TRACE
SDA> SET OUTPUT/NOINDEX trace.txt
SDA> SHOW TRACE
SDA> IO UNLOAD

Volker.
Jan van den Ende
Honored Contributor

Re: What is -SYSTEM-F-HWM_STALL?

Malcolm,

from your Forum Profile:


I have assigned points to 38 of 42 responses to my questions.

(one stream is fairly old)

Maybe you can find some time to do some assigning?

Mind, I do NOT say you necessarily need to give lots of points. It is fully up to _YOU_ to decide how many. If you consider an answer is not deserving any points, you can also assign 0 ( = zero ) points, and then that answer will no longer be counted as unassigned.
Consider, that every poster took at least the trouble of posting for you!

To easily find your streams with unassigned points, click your own name somewhere.
This will bring up your profile.
Near the bottom of that page, under the caption ├в My Question(s)├в you will find ├в questions or topics with unassigned points ├в Clicking that will give all, and only, your questions that still have unassigned postings.

Thanks on behalf of your Forum colleagues.

PS. ├в nothing personal in this. I try to post it to everyone with this kind of assignment ratio in this forum. If you have received a posting like this before ├в please do not take offence ├в none is intended!

Proost.

Have one on me.

Jan
Don't rust yours pelled jacker to fine doll missed aches.