Bug 112915
| Summary: | LTC5726-Network scripts report error when starting/stopping a TR device. | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 3 | Reporter: | IBM Bug Proxy <bugproxy> |
| Component: | anaconda | Assignee: | Jeremy Katz <katzj> |
| Status: | CLOSED CURRENTRELEASE | QA Contact: | Mike McLean <mikem> |
| Severity: | medium | Docs Contact: | |
| Priority: | medium | ||
| Version: | 3.0 | ||
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2004-10-05 14:41:12 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
| Bug Depends On: | |||
| Bug Blocks: | 123574 | ||
|
Description
IBM Bug Proxy
2004-01-05 21:49:17 UTC
----- Additional Comments From khoa.com 2004-02-06 09:48 ------- I've put this on RHEL3 QU3 list as a Sev 3. IBM, In doing some recent testing for a different problem, I have noticed that we are not hitting this bug. I looked at the ifcfg-tr0 file and there was no TYPE= line. The interface will work just fine without it. Using Token Ring for TYPE without the quotes reproduced this bug as described, adding in the quotes made it go away. At most this is a cosmetic issue. In our test environment, we used kudzu to add/configure the interface. We need to better understand how the TYPE=Token Ring got there in the first place. Did you use redhat-config-network by itself to configure the interface (no other tools or hand-editing of the file)? Also, please make sure this is reproduceable on U2 (or the U3 beta) I think the problem lies in r-c-network in tokenringhardware.py. Near the end of the script, we set self.hw.Type = 'Token Ring'. This eventually gets written out as "TYPE=Token Ring" in ifcfg-tr0. The fact that "Token Ring" isn't quoted makes bash think Ring is a command later on. I see 2 ways to fix this; condense Token Ring to 1 word (this is why all the other interfaces don't have this problem AFACT) Fix whatever writes the ifcfg files to quote whatever is after "TYPE=" I am setting this to all platforms as this is a platform independant bug. which version of rhpl is involved here? rhpl-0.110.6.1 is what I believe I was using. I can try it again, but it will be a little while until I have access to something with a token ring card . It appears that r-c-network won't write anything out so that I could see if the problem is still there or not unless the modules loads. No problem with # rpm -q rhpl rhpl-0.110.4-1 and this has not changed since... Steps to Reproduce: 1. Network install over a token ring device. 2. Boot the system I don't think anaconda writes TYPE= # cat /etc/sysconfig/network-scripts/ifcfg-tr0 DEVICE=tr0 BOOTPROTO=dhcp ONBOOT=no This file is enough to experiment with r-c-network... Just start it and the "Save". # cat /etc/sysconfig/network-scripts/ifcfg-tr0 DEVICE=tr0 BOOTPROTO=dhcp ONBOOT=no TYPE='Token Ring' I see no problem here.. Maybe a bug in anaconda/network.py, which does not add '' This was included in RHEL3 U3. network.py revision 1.24.6.14 date: 2004/08/03 03:13:19; author: notting; state: Exp; lines: +1 -1 token ring fix here as well (#128586) ----- Additional Comments From rayda.com 2004-10-05 13:15 EDT ------- I see no such messages in the boot up of the shipped QU3. My token ring works fine. Closing. ----- Additional Comments From rayda.com 2004-10-05 13:17 EDT ------- Sheesh, there's a lot of states to get through to find the CLOSED option.... |