All I did was remove a tape drive and then add a new tape drive to a SCO Unix machine. The process went fine. After testing the new tape drive I discovered I cannot get to the SCO Unix computer from any where else on the network. I cannot even ping it any more or ping any thing from it. The ifconfig -a looks o.k., atleast what I know to look for. The only other thing I am suspect about is shutting down the machine. It is restarted every night, but this is the first time in a while that it has actually been shutdown since installing a Windows 2000 server as the Domain Controller on the network.
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations derfloh on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
SCO Unix no longer on network - cannot determine why 2
- Thread starter GraceT
- Start date
-
1
- #2
Watch the startup messages closely for any errors. After it is booted, run
# hwconfig -h
and identify the network adapter. If it is listed, it normally has its MAC address listed as well. This is a good sign that the card is being identified by the driver.
When you swapped tape drives, did you have to relink the kernel to accomodate a different drive?
Check your hosts file entry for your own machine. Make sure that entry is valid.
Can you ping your own address locally?
(Stupid question below - Sorry)
Is it plugged into the network?
# hwconfig -h
and identify the network adapter. If it is listed, it normally has its MAC address listed as well. This is a good sign that the card is being identified by the driver.
When you swapped tape drives, did you have to relink the kernel to accomodate a different drive?
Check your hosts file entry for your own machine. Make sure that entry is valid.
Can you ping your own address locally?
(Stupid question below - Sorry)
Is it plugged into the network?
- Thread starter
- #4
I did an ifconfig -a and in the net1 listing was: ether 00:04:ac:b9:d9:f3. So I am assuming this is just as good as the hwconfig -h, but I will do that as soon as I get to the office. I am on my way as soon as I finish this reply.
I don't know if I had to relink the kernal, but I did. (It asked.)
Possibly the entry in the hosts file is incorrect because of the change from a Windows NT PDC to a Windows 2000 Active Domain Controller. I will check this too.
I can ping the address and I can ping the loop back address.
No question is stupid. I've done plenty of stupid things before. But yes, it is plugged in and there are lights. I even tried a different cable that goes though a different switch.
The IPs are assigned statically. According to the ifconfig -a, it has the address it always had, and the same subnet. It also reads <UP, BROADCAST, RUNNING, MULTICAST>.
- Thread starter
- #5
Problem solved! I did two things at the same time so I am not sure which solved the problem. Who knows, maybe I had to do both.
First of all I downloaded a new driver for my network card from ftp.sco.com/pub, then drivers, then 5.0.5, then network. Then I used the custom and netconfig commands to remove the network card and reinstall it with the new driver. The network card is a PRO/100+.
Then in the hosts file the machine was listed twice. Once apparaently from the old NT domain and then again for the new Win2000 Active Directory domain. I removed the old entry.
After I restarted everything was back to normal!!!!!
I used a lot of "I's" in this. Actually, the advice came from an advisor Pcunix at Or you can check out his web site at This second one has info and resources for Unix and Linux Systems and it is free. You have to pay to use his help on the first site I mentioned, but it is worth it if you are in a bind.
First of all I downloaded a new driver for my network card from ftp.sco.com/pub, then drivers, then 5.0.5, then network. Then I used the custom and netconfig commands to remove the network card and reinstall it with the new driver. The network card is a PRO/100+.
Then in the hosts file the machine was listed twice. Once apparaently from the old NT domain and then again for the new Win2000 Active Directory domain. I removed the old entry.
After I restarted everything was back to normal!!!!!
I used a lot of "I's" in this. Actually, the advice came from an advisor Pcunix at Or you can check out his web site at This second one has info and resources for Unix and Linux Systems and it is free. You have to pay to use his help on the first site I mentioned, but it is worth it if you are in a bind.
-
1
- #6
Glad to have been of help - I'm not sure what your problem really was either, but it never hurts to have the right driver.. so I figured we'd start there. And of course duplicate entries in hosts aren't helping the machine any, so clearing those out is just good housekeeping..
But WHY you were working before this and not after, I can't say from here. Several things I do want to point out though:
There was no need to do anything at all because of the new tape drive as long as it had the same scsi id as the one you took out.
The network scheme is wrong, wrong wrong and should be fixed.
You need to have recoverable backups:
When relinking kernels, two habits will save you much grief:
first, cp the existing /stand/unix somewhere safe, preferably on /stand (unix.good) if you have the room (see ).
Second, do a cd /etc/conf/cf.d;./link_unix before you do anything - just to make sure your current config will link, then say N to installing it but do compare sum of it to your working kernel to see that it is the same - if it's not, you don't have the same config as what you are actually running and really need to know why..
Best of luck, glad I was available for you, now back to my taxes
Tony Lawrence
SCO Unix/Linux Resources tony@pcunix.com
But WHY you were working before this and not after, I can't say from here. Several things I do want to point out though:
There was no need to do anything at all because of the new tape drive as long as it had the same scsi id as the one you took out.
The network scheme is wrong, wrong wrong and should be fixed.
You need to have recoverable backups:
When relinking kernels, two habits will save you much grief:
first, cp the existing /stand/unix somewhere safe, preferably on /stand (unix.good) if you have the room (see ).
Second, do a cd /etc/conf/cf.d;./link_unix before you do anything - just to make sure your current config will link, then say N to installing it but do compare sum of it to your working kernel to see that it is the same - if it's not, you don't have the same config as what you are actually running and really need to know why..
Best of luck, glad I was available for you, now back to my taxes
Tony Lawrence
SCO Unix/Linux Resources tony@pcunix.com
- Status
Similar threads
- Locked
- Question