Netware 6.5 error -634

Well, what you did wouldn't cause this behavior - no how, no way, period.

Looks like you've got some issues with your file system.  Are your disks setup in NSS or traditional?  I'm assuming they're NSS.

Does it totally hang completely or can you type something in to try and get the server running?

Yes, I can type commands.

Ok...I'm not sure what loadstage you are making it through, but what happens if you simply issue a "mount all" command?

At the prompt, is your server name even listed?  Or doesn't it get that far yet?

After "mount all"  it tells me that "data" is active and "data2" (another volume) is active.

Ok, so...can you see the files on them then?  I'm assuming you previously could not see them, but after running that command it mounted everything successfully?

If you do a VOLUMES at the System Console, can you see all of them?

If you type in EDIT at the SC, can you browse your files?

Yes I see the volumes and I can see all the directories and files in "Data"

Restart the server again and tell me what happens then.

OK  it is hanging at the same point with the message: Admin Volume Startup:" error logging the server in =-634  etc. . "  same message as my initial problem.

I may add that at System Console it is saying:' unable to communicate with Mail server.  "Mail" is a second physical server we have and that one is visible on the network. I can also add that we are not using it as a mail server anymore having switched from groupwise to google for schools just las week.

Then let's find out what it's hanging on.

When Netware loads, you can interrupt the boot process before the server actually launches the Netware operating system for maintenance.  There will be instructions right on your screen for interrupting the boot process.  Interrupt it.

Then type the commands I have listed here until the server hangs.  Tell me where it's hanging:

 STARTUP.NCF
 LOADSTAGE 1
 LOADSTAGE 2
 LOADSTAGE 3
 LOADSTAGE 4
 AUTOEXEC.NCF
 LOADSTAGE 5

There should be something on your system console or logger screen that shows a hung module or something the server is trying to do but cannot.

I see your error in what you are describing, and it's kind of hard to pinpoint what's causing it yet.  I just want to see if there's something that the loadstages will tell me.

Of course, since I tell you that it's not in any result of what you did...it got me thinkin'...you didn't accidentally delete important items in your autoexec.ncf file, did you?

It should be pretty large and have mount commands for your volumes and everything.  I don't know how familiar you are with it, but can you look at it?  A simple EA from the system console will open it and you can look it over.

I'd ask you to post it, but I'm sure you can't connect to it with a client.

My comment should have been "disabling DHCP services and REMming it out from autoexec.ncf will not yield your behavior" :D

Ok,
after startup.ncf  I get 7 or 8 modules "unresolved"  such as CPQSHD>CDM and ACPIMGR.NLM and more.
on loadstage 1 I get: "Critical component CONNMGR.NLM is not loaded or missing."
It does not go on to stage 2.

Look at your autoexec.ncf.  Does it look like all your information is in there?

I see 4 autoexec:  .bat  .bak   .old  .org  but they are just a few lines long!  What I did was add a semi colon in front of two programs I did not want to run . . .

You are missing your autoexec.ncf?

Well...that's not good then.

See what your other ones look like as far as how close they were to your existing one, like the bak or the old one.

You may have to see if you can simply restore it from tape and copy it to a floppy or something.  Or if a backup one has enough information, that would get your server running, depending on how close it was to the last one.

So doing an EA at the system console prompts you to create a new autoexec.ncf file?

Something definitely went wrong when you saved it :(

Oh, I see you wrote that they are just a few lines long...

Let me ask you this - do you make backups of your system and save them to a volume when you service pack a server?

If so, you might be able to find the old one since your last system update (if that's applicable)

For instance...you could look in sys:system\backsp***\system\ and find your autoexec.ncf in there.  Check it out once and see.

Else you might have to see if you can grab it from tape.  Worst case is that you can copy it back into place or re-type it if you can view it on a screen.

OK,  I would swear I saved the ncf before closing. . .  Now two questions:
One is: can I copy data from the server to an external drive from console commands?
Two:  I have a backup tape in the server with probably the info on it, but I don't know how to access it from the console.  I don't know its name as I only used it through the Veritas app on a remote workstation.

Read my previous question before you go that route.

Depending on your version of BackupExec, if you could actually communicate and see all your volumes and load all your modules (which you won't be able to) you would be able to do it.

But that's why I refer you to my previous question.  See whether or not you have AN SP backup, because you'll have your file back in no time then.

And chances are you won't be able to do anything with plugging in an external drive if you only made it through loadstage 1 to copy data :(

Well, I can't find a backup file. . . Basically all I see is the C: Drive of the server.  There are a few bakcsp2, 3 etc. in there but they don't seem to contain data. . .  or only Drivers. . .      Isn't there a "deleted files" folder where you can retrieve deleted files? It seems to me I was able to retrieve some data that someone had accidentally deleted once. . .

Yeah, but not without the volumes being mounted.  You said you could see your volumes before though, right?

Can you try a mount all command again and see if you can get them going again?  I thought you were that far along before.

OK,
I spent some time looking carefully through the directores and I found an autoexec.ncf in the backup tape which I could access.  Now what do I do?  I see the file and could type it down and copy it exactly but is there a way to copy, move. . .?

Do you have a floppy drive on that server?

Yes I do

I'm at the dentist so I have no links.

Grab toolbox.nlm from a google search which is easy to find. Copy the nlm to a floppy and copy the autoexec.ncf contents or file itself to the floppy drive as well.

Load toolbox.nlm and use its copy function to put the file back. It may work without authentication and hopefully with your current boot progress. Else, worst case is you are retyping by hand.

Somehow I am unable to load toolbox from the floppy.  I tried to copy it to the system folder but it said I don't have the right authentication to access the server tree. . .
I guess, I am stuck with retyping carefully every line. . .  which is better than losing everything!
So if I choose create an autoexec.ncf  and type in the information, that file will be in the right place automatically?

Create the autoexec.ncf file in the sys/system folder. If there is a way to copy it with netbasic or something I don't know how off the top of my head.

No need to retype it all - just disk and nic commands after the server info. Then the server will come up and you can connect to it after it mounts the volumes and activates the nic. Then copy/paste the rest.

Just to be clear:  any line preceded by a pound sign in the autoexec is actually NOT a command line and can be skipped, or?
There are too many lines I don't particularly identify (the NIC one is easy) the disk info is less clear.  .

Wow!  Thanks a lot for staying with me most of the day and patiently answering all my questions. . .  We are up and running now, and it feels good to come back from the abyss!
Again, thank you, you are a wonderful, helpful and articulate tech help.