202797 2007-12-19 15:58 0000 rox-base/rox-session: a minor issue 2008-02-21 13:52:27 0000 1 1 1 Unclassified Gentoo Linux Applications unspecified All Linux RESOLVED FIXED P2 normal --- 1 galtgendo@o2.pl rox@gentoo.org znmeb@cesmail.net galtgendo@o2.pl 2007-12-19 15:58:46 0000 I don't know, if it's a bug. If it is, I don't know if it's in rox-session, rox-lib or dev-lang/python' logging.py. For quite awhile (over a year) I've been getting traceback at rox-session start. It did not prevent rox-session from running, was annoying nevertheless. Lately I've decided to ask a question about it on rox-devel. I did not received a full answer, but something came up. I was advised to change in logging.py msg = msg % self.args to msg = repr([msg] + self.args) That was giving a traceback too, but when I change it to msg = repr ([msg].extend(self.args)) traceback was gone and the message printed was WARNING:root:None So does anybody have an idea if it's a bug, and if so, which package gives it ? As a sidenote, does anybody know a way to view all sourceforge lists online, cause the problem I had with rox-devel happens on other lists too, and not all are on gmane (at least, I don't think they are) ? znmeb@cesmail.net 2008-01-04 04:39:07 0000 I'm just starting to test ROX -- you seem to have gotten further than I have. I can't even get the window manager to come up. See bug 204239 lack@gentoo.org 2008-01-04 21:44:45 0000 > Lately I've decided to ask a question about it on rox-devel. > I did not received a full answer, but something came up. Yes, I saw this there - Very curious, especially your latest result incorporating Thomas's latest suggestion. What does the following command return when on your system: $ grep 'failed:' /usr/lib/rox/ROX-Session/*.py I hope your answer *should* rule out any bug in the way Gentoo installed ROX-Session. galtgendo@o2.pl 2008-01-04 22:57:23 0000 Well, it returns: /usr/lib/rox/ROX-Session/xsettings.py: warn('%s failed: %s', (cmd[0], exc)) galtgendo@o2.pl 2008-01-04 23:01:30 0000 And (I failed to mention it in upstream post) same warning as for xset is printed for xgamma (I got it after I removed logging.pyo, so that's why it's missing there). lack@gentoo.org 2008-01-18 13:43:41 0000 Here's a useful fix: I'll just make rox-session include xset and xgamma. I know this isn't *really* a fix of the logging issue, which should be resolved by upstream. But it should get rid of the errors if xset and xgamma are available for rox-session to run. Once the changes hit the rsync mirrors, reinstall rox-session-0.30-r3 (Check first to ensure that this reinstall will install xset and xgamma). Or alternatively, you can just run this: emerge xset xgamma Hopefully that will make things better on your system. I'm not sure what I can do about the other logging issue, but I'll keep this bug open until it is solved. galtgendo@o2.pl 2008-01-18 14:42:29 0000 Well, the catch is, though xgamma is not on my system, xset is. More details: at this time (while xgamma is not emerged) warning for xgamma is: WARNING:root:xgamma failed (so OK) for xset it's the one with `interrupted system call` and it's printed twice, so this may be some race condition. So I don't think emerging xgamma will fix this. So this are two bugs: one for logging.py and one for this 'possible race condition or something' galtgendo@o2.pl 2008-02-14 13:40:21 0000 If, for some reason, somebody is still following this bug, you should know that the traceback part has been solved on sourceforge list (too minor for a new release). It was a simple typo in one of rox-session files. The 'errno 4' part is still there, my uneducated guess is it's some kind of race condition, but that doesn't bug me as much as the traceback did. lack@gentoo.org 2008-02-21 13:52:27 0000 Okay, I have incorporated upstream's logging patch into rox-base/rox-session-0.30-r4 Enjoy!