[00:00] am i allowed to post a link to my site that is now running? [00:00] unimatrixoverlod: glad you got working [00:00] unimatrixoverlod: not sure that's ontopic or appropriate for the support channel [00:00] *support/discussion [00:01] nacc: got it. no problem. thank you Nacc again for your help. [00:01] unimatrixoverlod: np! [00:01] couldn't have done it without you :) === skylite_ is now known as skylite === gaughen_ is now known as gaughen [03:14] is it possible using filezilla or some other client to upload an ascii file to one server and a binary to another server, without corruption between the two? [03:14] (FTP) [03:36] I'm currently trying to fix someone's mail server and it's having quite a strange issue. The server is using dovecot and for some reason incoming mail is going to /var/mail/%n, instead of /home/mail/%n, after a lot of attempts at changing various things over and over to make them point to mail to the /home path [03:36] Ubuntu 16.04.1 LTS [03:37] i have a strange issue with tcpdump not wanting to read files, or write files in certain directories.. chmod ugo+rw is set for that directory, single partition, i can dd if=/dev/random into that file, running tcpdump -Z root didnt help.. sudo, or running it as root makes no difference [03:39] mkdir /test; cd /test; tcpdump -w /test/whatever [03:39] open("whatever", O_WRONLY|O_CREAT|O_TRUNC, 0666) = -1 EACCES (Permission denied) [03:40] umask 022 [03:41] setuid(0) setgid(0) too [03:41] writing a file to /tmp and moving it to that directory, and i get same thing permission denid [03:43] app armor? [03:46] audit: type=1400 audit(1474602137.806:48): apparmor="DENIED" operation="mknod" profile="/usr/sbin/tcpdump" name="/test/whatever" pid=20916 comm="tcpdump" requested_mask="c" denied_mask="c" fsuid=0 ouid=0 [03:46] not sure why i didnt run into this earlier [04:39] can anyone recommend a really simple monitoring system>? [04:39] nagios is overkill and has too many problems [04:39] like NPRE or whatever it's called having massive security issues [04:39] all i want to monitor is disk, cpu and security updates on 3 servers [08:04] Hello! Is anyone willing to help be debug a strange Apache rewrite problem that returns either Error 500 or 301? === _degorenko|afk is now known as degorenko [08:38] Could anyone help me figure out this problem, please: http://stackoverflow.com/questions/39611923/apache-rewrite-adding-unnecessary-directory-get-returns-error-500 ? === jamespag` is now known as jamespage [12:55] ddellav, coreycb: https://objectstorage.prodstack4-5.canonical.com/v1/AUTH_77e2ada1e7a84929a74ba3b87153c0ac/autopkgtest-yakkety/yakkety/amd64/p/python-pbr/20160923_035406@/log.gz [12:55] autopkgtest failures for pbr [12:55] :~( [12:55] jamespage, grr [12:58] jamespage, python-wheel is in universe [12:58] coreycb, I know [12:59] coreycb, its only in the tests afaict [13:00] jamespage, yeah I"m only seeing it imported in that one test file [13:01] jamespage, maybe just skip TestPackagingWheels() [13:02] coreycb, oh wait there is a --derives-from in the rules [13:02] the package skips unit tests - this is a main/universe re-org hangover [13:02] coreycb, you can add python-wheel to the autopkgtest depends - that's allowed [13:02] jamespage, ah, cool [13:02] didn't know that [13:03] coreycb, actually if you add it to the BD's and stop skipping the tests during the pkg build that's also good [13:03] coreycb, but we should minimize the cleanup now I guess [13:04] coreycb, the autopkgtests use the BD's so I'd add python{3}-wheel to the BD's for the package [13:05] jamespage, ok let me try that [13:26] jamespage, this is new for me, there's c code in the pbr tests [13:28] oh I think I just need python-all in BDs [13:40] cpaelzer: FYI, I figured out bug 1620955 [13:40] bug 1617963 in heimdal (Ubuntu) "duplicate for #1620955 package libasn1-8-heimdal:amd64 1.7~git20160703+dfsg-1 failed to install/upgrade: package libasn1-8-heimdal:amd64 1.7~git20160703+dfsg-1 cannot be configured because libasn1-8-heimdal:i386 is at a different version (1.7~git20150920+dfsg-4ubuntu1)" [Undecided,Incomplete] https://launchpad.net/bugs/1617963 [13:49] rbasak: was it starting to show up more often? [13:50] rbasak: did you spot that by the versions they had reported? [13:51] rbasak: or by running into the same issue [13:52] cpaelzer: yeah it seemed that three people hit it separately, so I looked at it a bit deeper and spotted that the versions are all from yakkety-proposed. [13:53] rbasak: you might want to tell him that he did help, that last comment appears to me that he considers your response as too-hard [13:53] last comment 2 minutes ago [13:54] cpaelzer: I'm working on answering this FAQ. [14:02] cpaelzer: with bug 1625577, I feel that it should either be marked with the "not enough info" template, or added to our backlog, but not left hanging. Otherwise if there's no activity the user doesn't have any direction on what to do. [14:02] bug 1625577 in php5 (Ubuntu) "package php5-mysql 5.5.9+dfsg-1ubuntu4.19 failed to install/upgrade: subprocess installed post-installation script returned error exit status 10" [Undecided,New] https://launchpad.net/bugs/1625577 [14:03] Well, I was punting the decision if it is not enough info or just my lack of php-wisness to nacc [14:03] which is why I have subscribed him and mentioned that in the comment [14:03] backlog is probably right [14:04] since it is not excluding my former intention [14:04] and avoids that it gets lost [14:04] thanks for letting me know [14:04] * cpaelzer is subscribing === Guest4978 is now known as IdleOne [14:26] ddellav, hey can you push your upstream and pristine-tar branches for ironic stable/mitaka release? [14:46] ddellav, looks like most of your mitaka branches need to get rebased. [14:48] Thanks [14:49] ddellav, sahara's an interesting one since we sync that from debian now, we probably should maintain ubuntu/* branches in alioth. [14:50] i was able to boot from maas 1st server client from pxe but i have many quastion lead by 1st pxe client boot with dedicated machine (not kvm/pxe boot ) === JanC is now known as Guest73904 === JanC_ is now known as JanC [15:09] ok i will br right back [15:11] coreycb is the mitaka staleness due to it sitting around for over a week or did i miss something originally? [15:12] coreycb also did you figure out pbr? It definitely ran tests during my local build [15:12] ddellav, just staleness [15:13] coreycb ok, i'll rebase them all and figure out what happened with ironic [15:14] ddellav, still working on pbr. it doesn't run unit tests during builds for ubuntu. the problem is autopkgtests (see debian/tests) are failing, they get run before a package can get promoted out of proposed. === iberezovskiy|off is now known as iberezovskiy === iberezovskiy is now known as iberezovskiy|off [15:58] jamespage, pbr tests are hitting something similar to this old bug https://bugs.launchpad.net/pbr/+bug/1296200 [15:58] Launchpad bug 1296200 in PBR "Unit tests are failing in Sid" [Medium,Fix released] [15:58] gpg: invalid item '(setpref)' in preference string === nacc_ is now known as nacc === degorenko is now known as _degorenko|afk [20:22] How do I configure my server to send all system mail via an external SMTP server [20:24] The eventual goal is to get psad to send alerts through this other SMTP server, which it seems like I can do by getting /usr/bin/mail to do it [20:24] Google's telling me that I would need to configure a whole new postfix daemon, and I'd really rather not go through all that just to tell my box where my mail server is [20:25] yes, that's what you should do. and it is easy, just answer a couple of debconf questions saying that it is a satelite system with a mail server that it should send everything to [20:25] jerichowasahoax: you can install something like ssmtp or msmtp (which just either deliver mail or fail, no queue) or nullmailer or dma (which are lighter/simpler than a full MTA but do include a queue) [21:00] something with a queue is probably a good idea if you don't want to lose any mail... [21:08] * tarpman generally installs nullmailer everywhere [23:00] hello, just installed server in a VM with LAMP, able to hit apache default splash, setting up dreamweaver SFTP connects successfully but cannot write to dir using default user account ubuntu created during install [23:01] do i need to create a special user / user group or modify the permissions of the folder? whats the common practice here [23:03] !chown [23:03] An explanation of what file permissions are and how they can be manipulated can be found at https://help.ubuntu.com/community/FilePermissions [23:03] which directory are you trying to write into? which user account owns it now? does that user account make sense? or would it make more sense to change the owner to the ubuntu user? [23:05] id assume there an existing group already and I need to add the default user account created during setup to that group? [23:08] lol, /var/www/html belongs to root, oh boy... [23:09] thanks for the rabbit hole [23:09] 'root' makes a certain amount of sense; afterall, root owns the usual web ports :) [23:09] but if you'd rather ubuntu own it, go for it. [23:12] well, id rather create a "webdev" group , add the default created user to install to that and then give that group write to /var/www/html/, is that ok or "not normal" ? [23:12] that sounds much better [23:14] deadhead: investigate 'bsdgroups' mount option (mount(8) has a short description) -- if you set the setgid bit on the directory, that will help keep files and subdirectories owned by the group, to make it easier for group members to work in the directory [23:18] sarnold, thanks, i was actually looking at just chgrp -R and chmod -R and be done with it but I am not a linux person and dont know if this removes the root group from the dir and would cause problems [23:18] ill dig before going forward [23:18] it would remove the root group, it's unlikely to cause problems [23:21] sarnold, do you know if /var/html/www is usually changed to 770 or 774 or there isnt reason to execute anything in here w 775 right? [23:23] deadhead: 775 is nice if the web server doesn't own the data and isn't part of the group -- that way it can get read-only access to the data via the 'world' bits [23:25] like a hosters web server of many clients, I see [23:27] principle of least authority -- I like to have as few resources modifiable by processes that interact on the network [23:31] so much to learn, we are strictly windows asp .net over here but jumping in the deep end for a swim, thanks for the help [23:33] woot, welcome aboard :) === mrx is now known as Guest56975 === Guest56975 is now known as eaux