Daniel,

Since you were able to use your IPL tape to do the initial installation, it
should still work to help you out here.  IPL from the tape, insmod the
dasd_mod module with what ever DASD device numbers you need, mount your
volumes, and download what ever RPMs you need to the system.  If you mount
your root file system on /mnt, and then your other files systems on
/mnt/whatever, you should be able to do a "chroot /mnt" command, and then
use RPM to install/de-install/whatever to get yourself back to your previous
condition.

Mark Post

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
Sent: Thursday, June 13, 2002 9:40 AM
To: [EMAIL PROTECTED]
Subject: upgraded rh 7.2 kernel, lost ctc connection (no network
connectivity)


Running rh 7.2 in a test LPAR.  I recently upgraded my rh 7.2 kernel
with rpm -Fhv kernel-tape (from 2.4.9-17 to 2.4.9-37).  Everything came
up fine except my ctc connection :(.  I had made tar backups of boot,
and of /lib/modules/kernel-tape2.4.9-17, but I placed the backup of
/lib/modules/kernel-tape in that directory.  When I rpm -Fhv
kernel-tape* the directory (and backup) got removed.  Foolish of me.

linux is tcsl, and the ctc gateway is tcsm (172.16.18.129).  It's always
been a little finicky, I couldn't just reboot linux.  I had to go down
to the hmc, send "ifconfig ctc0 netmask 255.255.255.128 down" on tcsl,
send "vary tcpip,,sto,cdevtcsl" and "vary tcipip,,sta,cdevtcsl" on tcsm,
then back in tcsl "ifconfig ctc0 up" and "route add default gw
172.16.18.129".

Well, now here are the errors I get:
when starting up: CTC driver Version 1.54 with chandev support initialized
ctc0 read ch 1370 (irq 04fa) write ch 1371 (irq 04fb) proto 0
ch-1370 interface disc or sel reset (remote)
ch-1371 interface disc or sel reset (remote)

I get a
ctc0 connected with remote side
ch-1370 Interface disc or Sel. reset(remote)
ctc0: Got remote disconnect, re-initializing
ch-1370: Interface disc or Sel. reset (remote)
ch-1371: Interface disc or Sel reset (remote)
ctc0 error during RX init handhsake
ctc0 got remote disconnect, reinitializing
ctc0 timeout during TX init handshake
ctc0 TX channel restart

these keep on repeating, along with some retries.

if I bounce cdevtcsl I get:
ctc0: Error during RX init handshake
ch 1371 remote side issued Read?, init...
ch 1371: unit check zero
ch 1370: unit check ZERO
ctc0 got block length 2!=rx length 32768
Timeout during TX init handshake
ctc0: TX channel restart

etc etc.

/proc/subchannels does show my ctc devices.

Also, when I issue a ifconfig ctc0 down on the linux side, it's gone for
good.  no longer shows up in ifconfig, and ifconfig ctc0 up fails.

What can I do?  I have a feeling restoring /boot/ and re-rerunning zipl
won't work because I've lost /lib/modules/kernel-tape2.4.9-17.  I'm not
sure how to get the older kernel rpm back on there without network
connectivity.  I've really messed this one up.  Any thoughts?

~ Daniel

Reply via email to