Gary Thomas wrote:
Gary Thomas wrote:
Chris Lumens wrote:
Traceback (most recent call last):
File "/usr/bin/anaconda", line 986, in <module>
kickstart.processKickstartFile(anaconda, opts.ksfile)
File "/usr/lib/anaconda/kickstart.py", line 879, in
processKickstartFile
ksparser.readKickstart(file)
File "/usr/lib/python2.5/site-packages/pykickstart/parser.py",
line 675, in readKickstart
self._stateMachine (lambda: fh.readline())
File "/usr/lib/python2.5/site-packages/pykickstart/parser.py",
line 553, in _stateMachine
self.handleCommand(lineno, args)
File "/usr/lib/anaconda/kickstart.py", line 831, in handleCommand
KickstartParser.handleCommand(self, lineno, args)
File "/usr/lib/python2.5/site-packages/pykickstart/parser.py",
line 407, in handleCommand
self.handler.dispatcher(args, lineno)
File "/usr/lib/python2.5/site-packages/pykickstart/base.py", line
321, in dispatcher
self.commands[cmd].parse(args[1:])
File "/usr/lib/anaconda/kickstart.py", line 488, in parse
fsprofile = pd.fsprofile)
AttributeError: FC4_PartData instance has no attribute 'fsprofile'
This will be fixed in the next build of anaconda. Thanks.
Could you send me the patch(es)? I have to build anaconda
myself (other reasons because I'm working with a non-standard
embedded platform) anyway.
I picked up the latest code from GIT and rebuilt. Now, I get
a bit farther (no Python exceptions), but it fails on this
line in my kickstart file:
clearpart --linux --drives=sda
The message (which is not in the log BTW) is:
Specified nonexistent disk sda in clearpart command.
This drive is definitely there - see attached /proc/partitions
Any ideas? All I'm trying to do is get a headless replay of
an install I did a few days ago...
BTW, the attached patch (for libdhcp-1.99.5) gets around the
problem I had yesterday with the kickstart and repository both
being loaded via NFS. I'm not sure if this is the correct
place for the patch, nor even if this is the best way to solve
the problem, but it works.
Note: the problem is that the NFS fetch of the kickstart file
leaves the network running and trying to restart it via DHCP
when mounting the repository fails because of the left-over
routes. My patch works around this, but only when the route
being added is marked by the system as already existing. It
should be safe under all conditions.
I'm still looking at the 'clearpart' error reported earlier today.
Thanks
--
------------------------------------------------------------
Gary Thomas | Consulting for the
MLB Associates | Embedded world
------------------------------------------------------------
--- libdhcp-1.99.5.ORIG/src/nic.c 2008-01-03 14:00:02.000000000 -0700
+++ libdhcp-1.99.5/src/nic.c 2008-01-15 07:25:29.000000000 -0700
@@ -2545,18 +2545,25 @@
return NIC_FAIL;
}
-NIC_Res_t nic_add_route(IProute_t rt) {
+NIC_Res_t nic_update_route(IProute_t rt) {
if (rt == NULL)
return NIC_FAIL;
- return nic_route_msg(rt, "nic_add_route", "add", RTM_NEWROUTE, NLM_F_CREATE | NLM_F_REPLACE);
+ return nic_route_msg(rt, "nic_update_route", "update", RTM_NEWROUTE, NLM_F_REPLACE);
}
-NIC_Res_t nic_update_route(IProute_t rt) {
+NIC_Res_t nic_add_route(IProute_t rt) {
+ NIC_Res_t res;
if (rt == NULL)
return NIC_FAIL;
- return nic_route_msg(rt, "nic_update_route", "update", RTM_NEWROUTE, NLM_F_REPLACE);
+ res = nic_route_msg(rt, "nic_add_route", "add", RTM_NEWROUTE, NLM_F_CREATE | NLM_F_REPLACE);
+ if (res == NIC_FAIL) {
+ if (nl_get_errno() == EEXIST) {
+ res = nic_update_route(rt);
+ }
+ }
+ return res;
}
NIC_Res_t nic_remove_route(IProute_t rt) {
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/anaconda-devel-list
