Bug 658835
| Summary: | repeated attempts to set regulatory domain from cfg80211 | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Peter F. Patel-Schneider <pfpschneider> |
| Component: | crda | Assignee: | John W. Linville <linville> |
| Status: | CLOSED CANTFIX | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | medium | Docs Contact: | |
| Priority: | low | ||
| Version: | 14 | CC: | ckom26, linville, mcgrof, olivier.berger, simonlanzmich |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2011-01-17 18:20:46 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: | |||
|
Description
Peter F. Patel-Schneider
2010-12-01 12:49:58 UTC
If you stay in one place, connected to a single network, and without removing or adding hardware, does it stop calling CRDA? It is a bit hard to answer this question definitively. What generally happens (not that I've characterized all the behaviour patterns) is that there are several fairly closely spaced CRDA calls (as above) and then no calls for a while. However, after a while there are often CRDA calls again. This is all without moving the laptop or messing with its hardware, but there are "competing" networks in the area, so maybe a competing WiFi network is messing with something. (But then why would this trigger CRDA re-determination?) Anyway, the problem occurs frequently so I should be able to easily provide debugging output if I'm told how to set it up. I'm seeing the same behavior, but with somewhat worse consequences. After most (*) of the regulatory domain changes, my wifi fails to reconnect. However, the available wireless networks are still visible (**). Suspending to RAM and resuming again makes the wifi usable again — until the next domain update. (*) My wifi has a relatively low signal strength of about 40%, which somehow seems to be related. When I move my computer to some place of better signal quality, the chances to reconnect without suspeding in between are somewhat better. (**) In rare cases, the wifi seems to completely lose contact and lists none of the networks. I'm using Fedora 13 on a MacBook (ath9k wifi) produced for Germany but currently am in Sweden, so the regulatory domain changes are to SE. Kernel version is 2.6.34.7-61.fc13.x86_64, NetworkManager is 0.8.1-10.git20100831.fc13.x86_64. I'm getting this problem again. It looks as if it is making one of the wireless networks I can see impossible, as the connection attempt times out before the repeated domain changes finish. The problem is not completely reproducible, but I should be able to do some testing, if I'm told what to look at. I've seen someone else report similar things on linux-wireless. See this thread: http://www.spinics.net/lists/linux-wireless/msg62427.html Essentially my conclusion was that the device is simply disconnecting a lot. You may want to review iw event -t log and see why. Indeed, in the opening report each crda call is preceded by a disconnect/reconnect cycle. I think that the multiple crda invocations is the symptom, not the problem. In hopes of avoiding extra confusion, I'm going to close this "crda" bug. If problems persist, you should open a new "kernel" bug specific to your driver. Do you have a /etc/modprobe.d/cfg80211.conf stating options like : options cfg80211 ieee80211_regdom="EU" So, try to comment out that line, and see if this improves the situation. Oliver, your advice fixed the repeated crda calls for me. Cheers! |