Perl 6 - the future is here, just unevenly distributed

IRC log for #darcs, 2015-03-25

| Channels | #darcs index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:16 mizu_no_oto joined #darcs
00:53 favonia joined #darcs
01:01 siel joined #darcs
02:42 mizu_no_oto joined #darcs
03:28 lispy_ joined #darcs
03:51 ilbot3 joined #darcs
03:51 Topic for #darcs is now http://darcs.net/ | logs: http://irclog.perlgeek.de/darcs/ | darcs 2.8.5 is out http://darcs.net/Releases/2.8
04:16 mizu_no_oto joined #darcs
04:43 c74d joined #darcs
05:42 dino- joined #darcs
07:36 sm joined #darcs
11:13 Wu joined #darcs
11:13 Wu hi!
11:13 Wu is there any way to see why a patch depends on another patch?
11:13 Wu (like, lines of code that sets such a dependency)
12:18 mizu_no_oto joined #darcs
12:45 Riastradh joined #darcs
12:48 mizu_no_oto joined #darcs
14:15 mizu_no_oto joined #darcs
14:25 mizu_no_oto joined #darcs
14:29 maus3r joined #darcs
14:30 maus3r Hi, I have a problem with a locked repo on darcs hub and don't know how to remove the lock file. Could someone help pls?
14:42 sm maus3r: I can help, which repo ?
14:43 maus3r great thanks, repo is called MUNICH
14:43 sm what makes you say it's locked ?
14:43 maus3r error messages about the lock file when pushing
14:44 sm if you try it now, same thing ? I don't see a lock file
14:44 sm oh pardon me, I do
14:45 sm oh, maybe because you're testing right now
14:45 maus3r yes I was testing
14:45 maus3r and there was no error
14:45 maus3r strange
14:45 maus3r could you remove the events.lhe.gz file?
14:45 sm darcs hub removes stale lock files every so often
14:46 maus3r this is a mess, it was a test, which was no good idea
14:46 sm (hourly)
14:46 maus3r In fact I wanted to remove the large file and every related patch
14:47 sm maus3r: you can delete the repo and re-push a new one, perhaps after obliterating some patches locally
14:47 sm you can also obliterate patches on hub, there should be instructions in the FAQ
14:47 maus3r ok I will try with oblitering
14:48 maus3r last question: there is no way to delete the lock manually, is there?
14:48 sm at the moment, no you just have to wait until the hour turns
14:49 maus3r ok
14:49 sm I forget how a stale lock file can get left over - did you cancel an operation or something ?
14:49 maus3r yes I canceled a push
14:49 sm ok.. that's supposed to work but I guess sometimes not
14:49 sm actually, there is a way to clear a lock - delete the repo
14:50 maus3r but that is rather drastic no?
14:50 maus3r will not all other users have problems then ?
14:51 sm you could push it again right away, I doubt anyone would notice even if you have multiple users accessing it (most darcs hub repos don't)
14:52 sm I'm not recommending it as a routine thing, just saying
14:52 maus3r ok I see
14:52 maus3r ok again the problem when trying to obliterate
14:52 maus3r ssh maus3r@hub.darcs.net obliterate MUNICH Waiting for lock /home/darcsden/users/maus3r/MUNICH/_darcs/lock
14:53 maus3r ssh maus3r@hub.darcs.net obliterate MUNICH
14:53 sm do you have other people working with this repo right now ?
14:53 maus3r no
14:53 sm did you cancel something again ?
14:53 maus3r maybe I created another lock file in my test
14:53 maus3r at the very beginning
14:54 sm there's a darcs apply job running on this repo right now
14:54 sm probably a push you cancelled. Shall I kill it ?
14:55 maus3r yes pls
14:55 maus3r and remove the lock
14:55 maus3r then I try to obliterate and if that works, I will stop pushing so big files ;-)
14:55 sm done
14:55 sm how big is the file ?
14:56 maus3r ok great! seems to be working again!
14:56 maus3r Thank you!
14:57 maus3r oh, the file was ~140MB
14:57 maus3r it was just meant as a test, since I could not find any info on the page wether there is a quota for the repos
14:58 sm right.. I'll have to add something
14:58 sm that may be ok, just takes some time to transfer as you noticed
14:58 maus3r there is basically no quota as a friend told me, but I should not push large files
14:59 maus3r this was before my test ;-)
14:59 sm <sm> burp: repos of a few hundred Mb are probably ok. A file of that size, probably not
14:59 maus3r äh after
14:59 maus3r yes ok, we have usually no large files
14:59 sm feel free to test with this one, if it's a problem I'll let you know
15:00 maus3r anyway, things are fine now, file oliberated and repo running
15:00 maus3r thanks again
15:00 sm you're welcome. Do you mind if I ask what you're using hub for ? school, work, personal.. ?
15:02 maus3r Its for science
15:02 maus3r at the university of zürich
15:03 maus3r particle physics ;-)
15:03 sm cool
15:04 maus3r I already write all my papers and small programs with darcs for a while, but now we want to do something bigger, and see if that works
15:04 sm nice
15:05 maus3r so have a nice day, I may be coming back at some point though ^^
15:05 sm good, feedback welcome
15:39 sm why would du -s somedir and du -s ../*/somedir report very different sizes for somedir ?
15:59 lelit maybe you have ./somedir, ../foo/somedir, ../bar/somedir ?
16:01 sm no, there's only one somedir
16:02 sm I was checking darcs hub usage by repo. Extremely puzzling behaviour by du
16:03 sm cd users; du -s */* shows numbers that are too small
16:03 lelit uhm, then maybe you have ./somedir/somedir? what I meant is that "../*/somedir" is different from "./somedir"
16:09 dolio joined #darcs
16:11 sm it's not, I swear :)
16:13 sm to be concrete (since maus3r already mentioned the name):
16:13 sm darcsden@joyful:~/users$ du -sBM */* | grep MUNICH
16:13 sm 408Mmaus3r/MUNICH
16:13 sm darcsden@joyful:~/users$ du -sBM mau*/* | grep MUNICH
16:13 sm 555Mmaus3r/MUNICH
16:13 sm
16:14 sm varying the kind of wildcard you start with makes a difference in the size reported
16:15 lelit oh, I see, sorry for being dense :)
16:24 sm it's du that's being dense >:|
17:36 mizu_no_oto joined #darcs
17:51 sm I think it's due to hard linking - "du only reports the space used by each file once, so files with multiple names (hard links) are only added to the first-scanned directory's total"
17:59 sm that was it!
18:02 sm find MUNICH -type f -printf '%n %p\n' | awk '$1 > 1{print}'` shows about 50 hard-linked patch files, which account for the difference. Good to know when measuring darcs repos.
18:05 amgarching joined #darcs
19:00 c74d joined #darcs
19:22 favonia joined #darcs
19:34 amgarching joined #darcs
19:54 mizu_no_oto joined #darcs
20:21 amgarching joined #darcs
20:22 lambdabot joined #darcs
20:24 c74d joined #darcs
20:59 masse joined #darcs
21:05 favonia joined #darcs
22:22 mizu_no_oto joined #darcs
22:25 favonia joined #darcs
22:56 mizu_no_oto joined #darcs
23:34 mizu_no_oto joined #darcs
23:45 favonia joined #darcs
23:55 MaicoLeberle joined #darcs

| Channels | #darcs index | Today | | Search | Google Search | Plain-Text | summary