GWinter.nlm abends server often

Hello,

For the last Six months we have been experiancing problems with our GroupWise Web Access agent.  The server is running Netware 6.0

On the server is

GW web access agent 6.0.2
GW Internet agent 6.0.2
Groupwise MTA 6.0.2

about twice a month the server abends on gwinter.nlm.  There seems to be no pattern in date or time. Is it common for this NLM to fail so often?  Also, I need to configure the health.log file.  It has gotten so big I cannot even open it.  What utility do I use to configure this log file?
edrosieAsked:
Who is Participating?
I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

PsiCopCommented:
HEALTH.LOG is not part of the GroupWise system. Is this a Compaq? Some server health agent is logging to that.

NetWare 6.0.... what SUPPORT PACKS, if any?

GroupWise v6.0 SP2... just FYI, that is not the latest SP for GroupWise v6.0.

It is not recommended to run the GWWA and the GWIA on the same machine. Altho you appear to have these gateways in their own Secondary Domain, which is good.

What is the webserver for the GWWA?

Also, are you running these GroupWise agents in a Protected Address Space, and if so, is each in its own space, or are they all in one space?

General tips:

1) Check disk space - make sure the Volume where the queue directories for the agents reside (this should NOT be the SYS: Volume!!) has free space

2) Make sure that deleted files are not clogging the filesystem - the queue directories for the various agents should generally be flagged Purge Immediate (use ConsoleOne)

What is the EXACT ABEND?
ShineOnCommented:
Duplicate Question.  The one in the NetWare TA should be changed to a 20-point "pointer" question...
ShineOnCommented:
As I said in the one in the NetWare TA:  HEALTH.LOG is the NetWare 6.0 server health log, located in SYS:\SYSTEM.  It is cleared/viewed using NetWare Remote Manager (NoRM).  Also, in that Q, I reaffirmed PsiCop's request for the exact abend, and expanded that to suggesting the upper portion of the abend log through the stack trace be posted.
Why Diversity in Tech Matters

Kesha Williams, certified professional and software developer, explores the imbalance of diversity in the world of technology -- especially when it comes to hiring women. She showcases ways she's making a difference through the Colors of STEM program.

edrosieAuthor Commented:
Here is the error log.

I will use set command to delete purged files immediatley since there is little room on this server.

Webserver is Apache 1.3.20a


Server NFIGW-1 halted Saturday, July 2, 2005   9:42:03 pm
Abend 1 on P00: Server-5.60b: General Protection Processor Exception (Error code 00000000)

Registers:
    CS = 0008 DS = 005B ES = 0010 FS = 005B GS = 005B SS = 0010
    EAX = C28DC960 EBX = CB6C3EBA ECX = CA659DE4 EDX = FFFFFFFF
    ESI = 00000000 EDI = 0000000C EBP = CB6C36A4 ESP = CB6C369C
    EIP = C28B18A7 FLAGS = 00010286
    C28B18A7 FF12           CALL    near ptr [EDX]=?
    EIP in VSXL5.NLM at code start +000128A7h

The violation occurred while processing the following instruction:
C28B18A7 FF12           CALL    near ptr [EDX]
C28B18A9 83C404         ADD     ESP,00000004
C28B18AC C9             LEAVE
C28B18AD 5F             POP     EDI
C28B18AE 5E             POP     ESI
C28B18AF 5B             POP     EBX
C28B18B0 C3             RET
C28B18B1 6838000000     PUSH    00000038
C28B18B6 E8B5ECD312     CALL    THREADS.NLM|__CHK
C28B18BB 53             PUSH    EBX



Running process: gwinter         6 Process
Thread Owned by NLM: GWINTER.NLM
Stack pointer: CB6C3620
OS Stack limit: CB6A4C40
Scheduling priority: 67371008
Wait state: 3030070  Yielded CPU
Stack: --FFFFFFFF  (LOADER.EXE|KernelTempAliasesEnd+FFF)
       --00000000  ?
       --CB6C36C0  ?
       --0000000C  ?
       --00000000  ?
       --CB6C3EBA  ?
       C28A14CD  (VSXL5.NLM|VSXL5VwGetRtns+EE1)
       --C28DC960  ?
       --00000000  ?
       --CB6C36E8  ?
       --0000000C  ?
       --00000000  ?
       --CB6C3EBA  ?
       CA65E687  (SCCFA.NLM|FAClose+34)
       --CA9CEC40  ?
       --C28DC960  ?
       D5607A0E  (THREADS.NLM|__Must_Have_Three_One_Or_Greater+A)
       --C2A3EC80  ?
       --00000000  ?
       --CB6C3818  ?
       --0000000C  ?
       --00000000  ?
       --CB6C3EBA  ?
       CA6768BA  (SCCDA.NLM|DACloseDocument+FB)
       --C2A3EC80  ?
       --CB6C3EBA  ?
       --CB6C3808  ?
       --CB6A2140  ?
       --00000000  ?
       --0000000C  ?
       --CB6C3818  ?
       CA6767EF  (SCCDA.NLM|DACloseDocument+30)
       --CB6C36FC  ?
       --005B0010  (UMSSHIM.NLM|__NLM_BSS_End+ABF4)
       --005B0008  (UMSSHIM.NLM|__NLM_BSS_End+ABEC)
       --005B005B  (UMSSHIM.NLM|__NLM_BSS_End+AC3F)
       --CA6C005B  (SCCEX.NLM|EXDTAddCellFrame+696)
       --CA6CECE0  (SCCEX.NLM|EXDTAddCellFrame+F31B)
       FC0048CB  (SERVER.NLM|kMutexUnlock+67)
       FC0048D4  (SERVER.NLM|kMutexUnlock+70)
       --CA6CECE0  (SCCEX.NLM|EXDTAddCellFrame+F31B)
       --CA67E9A0  (SCCDA.NLM|DAGetErrorString+4814)
       --CC67EDE0  ?
       --00000000  ?
       --00000000  ?
       --CC5E9640  ?
       --00000000  ?
       --CB68F3E0  ?
       D55F0C73  (THREADS.NLM|free+F)
       --CB6A2AE0  ?
       --CB6A2140  ?
       --00000000  ?
       --CB6A2AE0  ?
       --CB6C37E8  ?
       --0000000C  ?
       --00000020  ?
       --CB6C379C  ?
       FC01A819  (SERVER.NLM|TcoValidateThread+11)
       --CB6C37B4  ?
       FC01BB8B  (SERVER.NLM|kGetThreadStackInfo+37)
       --CB6A2140  ?
       --CB6C37B0  ?
       --CB6A2AE0  ?
       --CB6A2140  ?
       --00000000  ?
       --CB6A2AE0  ?
       --CB6C3A54  ?
       --CB6A2AE0  ?
       --CB6A2140  ?
       --00000000  ?
       --CB6A2AE0  ?
       --CB6A2140  ?
       --00000000  ?
       --CB6A2AE0  ?
       --CB6C3A54  ?
       --0000000C  ?
       --00000134  ?
       --CB6C36DC  ?
       FC01A819  (SERVER.NLM|TcoValidateThread+11)
       --CB6C3808  ?
       FC01BB8B  (SERVER.NLM|kGetThreadStackInfo+37)
       --CB6A2140  ?
       --CB6C3804  ?
       --CB6C3808  ?
       --CB6C3A54  ?
       --CB6C36DC  ?
       D55FCB58  (THREADS.NLM|FreeQueryServicesList+2518)
       --CB6A2140  ?
       --CB6C3804  ?
       --CB6C3808  ?
       --C39B8220  ?
       --C39B8220  ?
       --00000000  ?
       --00000000  ?
       --00000001  ?
       --CB6C3A54  ?
       --0000000C  ?
       --00000000  ?
       --CB6C3EBA  ?
       CA6EDBF2  (GWCMC.NLM|CMC_SetMaxConversationThreadCount+492)
       
Additional Information:
    The CPU encountered a problem executing code in VSXL5.NLM.  The problem may be in that module or in data passed to that module by a process owned by GWINTER.NLM.
PsiCopCommented:
Hmmm....kinda non-specific, and I'm not pulling up anything in the Knowledgebase.

Honestly, my first suggestion is to tell you to update your GroupWise installation to at least SP4. The docs for SP4 state a number of WebAccess Agent ABENDs were fixed in that SP (altho they are not very specific about the nature of them).
ShineOnCommented:
Problem number 1: you are running on NetWare 6.0 SP1 (5.60b)  You should be on SP5.  SP1 was very buggy.  The first decent SP, unfortunately, was SP3, and 5 is very stable.

Problem number 2: you are running GroupWise 6.0 SP2, which should be at SP4, which fixed several abend issues in both GWIA and GWWA.

I hate to sound like Novell Support... but...

Apply the current support packs.  If the problem persists, let us know.

Experts Exchange Solution brought to you by

Your issues matter to us.

Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.

Start your 7-day free trial
PsiCopCommented:
You can download the latest SP for GroupWise from -->

Note that SPs are cumulative. You do NOT need to apply SP3 prior to applying SP4.

A GroupWise SP should be applied to the Primary Domain first, then to Secondary Domains. Not all Secondary Domains need to be updated at once, but the Primary should be first.

It would also be helpful to know the Support Pack level of NetWare itself. At the console prompt, enter --> VERSION
PsiCopCommented:
Ah. Sorry, I missed that. I see it now. Yes, ShineOn is correct, NetWare v6.0 SP1 was not the most stable thing.
edrosieAuthor Commented:
I was affraid the suggestion might be to upgrade SP's.  I get a little nervous about that.  We have 7 domains, 600 users on 8 remote locations getting about half a million emails a month.  This is also why I want to go to exchange.  From what I am learning about exchange it is abit more user (admin) friendly.  This will call for a meeting or two before I can impliment.  I will have to convince a lot of people that email will not go down.
PsiCopCommented:
Ummm....you have 600 users getting 500k E-Mails/month and you think that Exchange will be easier to use?

Or cost less?

News Flash: Moving to Exchange will easily DOUBLE your organization's costs per mailbox, per month - if not triple that figure. And there will be no increase in reliability, nor will your administration be any easier.
PsiCopCommented:
I also fail to see how moving to EX would make you less nervous than applying an SP to an existing product.

Further, it mystifies me why you think its a valid comparison to pit GroupWise v6.0 - which is about 4 YEARS old and was EOLed last year - against the shipping version of EX. If you want to compare, you should be using at LEAST GroupWise v6.5, or v7 (I'd expect to see v7 released in the next 30 days).

Finally, you've neglected your GroupWise install for at least 18 months (since you're 2 SPs behind) and your NetWare install for at least 2 YEARS (since you're 4 SPs behind). What makes you think that if you subject a EX environment to similar neglect that your results will be better? Is your comparison of a neglected environment to a new one at all valid?
edrosieAuthor Commented:
OUCH!!!

OK, I learned my lesson... I'll ask questions only and keep my concerns to myself.  Yes, we neglect upgrades.  For us Novel is a pain.  There has never been a change (since the days of 4.1) that has not caused us headaches.  We can have file servers go down and very few people complain.  E-mail goes down and the CEO is at my desk in less then 10 minutes.  Go easy on me, I am just trying to find what is best for our company.
PsiCopCommented:
Wow. That's incredible. I just finished rolling out NetWare v6.5 SP3, and the worst hitch was when a hardware RAID controller in one server failed during the post-SP reboot - and that's hardly NetWare's fault, is it? We also recently deployed GroupWise v6.5 SP4 - again, no problems.

A NetWare server going down around here generally happens due to hardware, like the motherboard that croaked in one of our field offices. We don't have very much of the OS dying on us.

When I hear of situations like you describe, my general thought is that the environment was not properly designed/installed, or hasn't been maintained well. For example, failing to partition the eDirectory tree across the 8 locations. Or running GroupWise with the message stores located on the SYS: Volume and none of the queue directories set to Purge Immediate. Or running 2+ year old code. Mistakes like those can cause constant problems.

And if you're having the problems you describe with an environment that should be reliable - and both NetWare and GroupWise should be, when properly managed - then I tend to think that:

1) the most cost-effective solution is to address what was done wrong
2) moving to a different and more-expensive environment won't solve the underlying issues, but will cost the organization more

And hence, I don't see where you think the business advantage will accrue from changing platforms.
ShineOnCommented:
Going to Exchange would seem to me to be eminently stupid.  No offense meant, it's just a fact.  You won't save a whit in administrative overhead - in fact you'll find it to be a lot more work that you apparently now think it would be.  Plus, your licensing 6.0 costs will be usurious.  Keep in mind the difference in licensing models, for both the OS and the email product.  You could upgrade to GroupWise 7.0 (might as well, 'cause for an installation your size you won't be doing anything - GroupWise or Exchange - without a good deal of planning first) and have it cost less than migrating to Exchange.  Exchange licenses, on Select Agreement with SA, is about $35 a year for 3 years, per user or device.  Plus the 2-3 grand for the Exchange Enterprise server license.  Plus $700-1000 per server for the Windoze server licenses.  Plus, unless you already have all of your users covered by Windows Server 2003 user/device CALs, another $20 or so per for that.  Plus, all of the infrastructure changes you'd have to make to put in an Exchange-compatible configuration of AD (yes, even if you already have AD in place, it might need to be redone to make it work for an Exchange installation.)  Plus the migration cost and effort.

The GroupWise 7.0 upgrade, on a CLA, would be less than a third of what an Exchange migration would cost you. license-wise.  Upgrading to NetWare 6.5/OES won't cost all that much, either, and you'll be able to install as many servers as you want without paying for more server licenses.  AND GroupWise scales well beyond what Exchange can handle - even Exchange Enterprise.

Applying SP's are not hard.  At all.  When you apply NW6SP5, for warm fuzzies you can skip updating of disk/lan driver files, if that makes you nervous.  Drivers are the most likely items to cause an issue when doing a NetWare SP.

If you are having any problems at all, you are *definitely* not going to fix them by NOT applying the support packs that fix those problems.  Look at it from a Microsoft perspective:  If you were on Exchange 5.5 on Windoze NT4, would you choose NOT to apply NT4 SP6A because you're nervous, even though your SP4 servers have been crashing daily?

The only thing I can think of that might seem easier about Exchange is that any dummy (again, no offense meant) can look at screen shots of the Exchange MMC and think "that looks kinda like the other Windows stuff I do at home" and further think it's going to be easy because of it.

PsiCopCommented:
And if you think I'm unsympathetic, allow me to draw a parallel.

Remember NT v4 SP2? The one that trashed NT servers right & left, and made the rest even more unstable than usual? What kinda response do you think you'd get in the M$ forums if you'd posted (a few years ago) that you were running NT v4 SP2 and having BSODs? People would probably be telling you to update to something after SP2, because SP2 had known problems, right?

So, here we are telling you that NetWare v6.0 SP1 had known problems, and you should be running something after SP1, and then when you mention that NetWare's been unreliable for you and you're considering switching platforms, you wonder at the reaction? I mean, seriously, here you are running code that's probably at least 3 YEARS old, and has known issues. You've made the CHOICE to do this, but now that its not working well, you want to use that as a basis to switch? Does that logic really track for you?
ShineOnCommented:
I ran a NetWare 6.0, GroupWise 6.0 shop and had it stable to the point where it would only "go down" if I took it down.  We had more e-mail problems caused by our ISP, by rogue RBL's, by our antivirus gateway, than we ever had with GroupWise/NetWare. But I always kept current on my SP's.   I am now in an Exchange shop, and I'm trying to get a migration to GroupWise through "the system" because we have had several occurrences of downtime, service slowdowns, and goofy, oddball problems in our Exchange email system.

I currently have a pair of NetWare 6.0 SP5 servers that have been up for 152 days and 72 days, respectively, and the last time they went down was for scheduled updates - and they were up solid for months before that.  If your NetWare servers fail consistently, it's probably because you're so far behind on maintenance.  If your GroupWise servers are on flaky SP levels of NetWare, they're not going to fare much better than your file/print servers.

I will grant you that NetWare 4.1 was kinda shaky (4.11 and 4.2 were both pretty solid, after a few SP's), 5.0 had its unpleasant moments but 5.1 was (and is) solid, (except for SP2 - which was a low point for NetWare SP quality assurance,) and NetWare 6.0 SP1 and 2 were no great shakes (which is all the more reason for you to get off SP1 and apply SP5, which IS very solid).  Remember, that during that same timeframe, Windows went through some very shaky times with their SP's - in fact, more people put off applying NT4 SP6a because of the problems with prior NT4 SP's than with any other support/service pack on any other platform that I can think of, so don't think for a moment that NetWare's the only platform that has had problems...  and don't think for a moment that only Microsoft has done any QA improvements in their service releases.
ShineOnCommented:
Another point to ponder:  There may be one of the 3 vulnerabilities found in Netware in the past few years, in the IP stack you're running your WebAccess and GWIA on.  If for no other reason, you should schedule the SP to take care of that potential security issue.

That's as compared to the more than 3 vulnerabilities found in Windows and other Microsoft products in the past month.   How often would you avoid applying Microsoft security hotfixes?  If you'd let your servers autoupdate, how nervous would you be that the next auto-applied hotfix will break your email?  I know I'd be a lot more nervous about that, than about applying NetWare 6.0SP5.  It only takes an hour or so, not counting the backup beforehand for warm-fuzzies, maybe less if you've got a relatively fast server...

What I'd recommend is first, update all of your NetWare 6.0 servers to SP5.  Once that's done, plan an update of GroupWise to SP4.  One of the nice things about applying a GroupWise SP is that almost all of it can be pushed from a central location - even the client updates - without using ZEN or SMS or LanDesk Manager or whatever...

Does that sound right, Psi?
ShineOnCommented:
Once you have your NetWare servers at the current SP level, wait a bit and you'll see how much more stable they are... and smile while your users thank you.  They may not complain about the occasional server going down, but believe me, they're not happy about it. They've just been trained not to complain.

Again from personal experience:  I started here about a year and a half ago, and found my NetWare servers and Novell clients and ZFD 3.2 all neglected, miserably behind in SP's - and causing problems for the users on a regular basis.  After getting things up to snuff, they just never go down and never cause problems any more. The users are much happier, and say so, not necessarily to me, but to each other - and the CEO hears about it.  This has also freed up my time to chase the never-ending Microsoft-related problems while researching and implementing ways to save the company money, which also gets noticed.

It's a lot nicer being noticed for good things than to be unnoticed for bad things not happening.
It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today
Groupware

From novice to tech pro — start learning today.