OK I figured it out. Big issue from a small mistake.
I deleted the vldb.DB0 and vldb.DBSYS1 (moved it really) and then
when I restarted the AFSDB's. (i.e. they were empty)
Then I restarted all the Fileservers and this time the FileServer2
booted saying it had 192.168.254.31, that was the IP in the /etc/
hosts file. (which I forgot to change)
once I changed that and did the vos syncvldb for each server it was
fixed.
whew!
Thanks for everyones pointers, that helped me though this, director
or indirect.
Cheers,
Pucky
On Oct 20, 2006, at 12:39 PM, Derrick J Brashear wrote:
On Fri, 20 Oct 2006, Pucky Loucks wrote:
Ok, so this is a little tricky but here goes.
We had 3 subnets.
[]
AFSDB1 had
192.168.254.21
192.168.130.21
AFSDB2 had
192.168.254.22
192.168.130.22
AFSDB3 had
192.168.254.31
192.168.130.31
Fileserver1 had
192.168.254.20
192.168.130.20
Fileserver2 had
192.168.254.11
192.168.130.11
Fileserver3 had
192.168.254.48
192.168.130.48
Because of the need to change the ip's in the corp network we
needed to change the ip's in the managment network. NOTE: the
internal production network stayed the same.192.168.254.0/24
AFSDB1
192.168.200.111 (was 192.168.130.21)
AFSDB2
192.168.200.112 (was 192.168.130.22)
AFSDB3
192.168.200.113 (was 192.168.130.31)
Fileserver1
192.168.200.121 (was 192.168.130.20)
Fileserver2
192.168.200.122 (was 192.168.130.11)
Fileserver3
192.168.200.123 (was 192.168.130.48)
Yesterday I was dealing with issues getting the 'fs' process
working on the Fileservers. What I ended up doing was shutting
down the 192.168.200.0/24 ip's and everything was fine. Then I
did a vos syncvldb and it was saying that a certain volume was
located on another server. (note earlier someone ran vos syncserv
without doing a syncvldb before hand)
Now today I ran,
vos syncvldb on all servers and vos syncserv on all server and it
worked. I have the 192.168.200.0/22 network shutoff.
Fri Oct 20 11:25:41 2006 File server starting
Fri Oct 20 11:25:41 2006 afs_krb_get_lrealm failed, using
office.h2st.net.
Fri Oct 20 11:25:41 2006 VL_RegisterAddrs rpc failed; The IP
address exists
on a different server; repair it
well, that implies the new address already exists, with a different
uuid. did you changeaddr wrong? does the vldb suggest the wrong set
of volumes are on that server?
_______________________________________________
OpenAFS-info mailing list
[email protected]
https://lists.openafs.org/mailman/listinfo/openafs-info
_______________________________________________
OpenAFS-info mailing list
[email protected]
https://lists.openafs.org/mailman/listinfo/openafs-info
