=== MarkMaglana_ is now known as MarkMaglana [06:13] Good morning [07:08] hey [07:09] i'm running ubuntu server in a chrooted sftp server which has a directory mapped to the users home dir which is a s3 bucket mapped by fuse s3fs, I can mount the drive and list the contents, I can write to a folder but when I try move a file into a folder I get opertion not permitted, I su to the user same error, I tried as root same error, I unmounted the s3 bucket and re mounted it. still nothing, [07:09] it did work fine till just before easter where it looked like the mount dropped so i remounted it but i just can't move from the users root dir to the folder, I can create files directly in folder [07:14] idle here a bit paul98 as US wakeup might get more volunteers [07:14] lotuspsychje: thanks, it's just really odd one :( [07:38] People : hi ! I've setup an ubuntu lts 18.04 server and it works fine but for the ntp daemon which doesn't start at boot time. Why is "systemctl enable ntp" not working ? [07:53] which one of you are volunteers [07:54] soulseeker: volunteers are everywhere, if you have a question please just ask in the channel [07:55] when using the netinstalker at the select software option why doesnt it prompt for software sekections like gnome [07:56] soulseeker: you are in the server channel here, are you going to install gnome on a server? [07:58] lotuspsychje it is a laptop but it does need to run server software [07:58] paul98 hi [07:59] were you using paul981 [08:00] lotuspsychje it all installed but grub2 [08:00] default install hoping to avoid problems [08:01] soulseeker: yes it was me [08:01] I couldn't remember my registered nick lol [08:01] did you get it working paul [08:01] nope, spoke to aws and it seems a mount issh [08:01] issue* [08:01] i've just restarted the server to see if that fixed it, I don't know alot about it [08:02] lotuspsychje was that you? aws [08:02] soulseeker: can you make a full sentence that makes sense? [08:02] it isnt real as far as I can tell paul [08:03] I've had to restart my machine, so I don't know if my question was asnwered or not :-( [08:03] -of amazon [08:03] lotuspsycheje are you the amazon program? [08:04] i rebooted it but still get same issue. something has gone wrong. [08:04] paul what were you trying to copy [08:05] it's just a csv file. it's just s3 bucket mounted via s3fs, i can touch a file directly into the processed/ but i can't mv from the root of s3 bucket into /proccessed as user root or the use who is sftp user [08:05] paul did you read the dune series [08:06] although I can list the s3 bucket contents of root and the processed/ [08:06] this worked fine, then suddenly it looked like the mount dropped, so i remounted it, but i can't write to the processed dir on the bucket [08:06] outside the mapped mount dir I can aws s3 cp to the bucket any file [08:06] paul touch it where [08:07] paul touch the file reminds me of icecat [08:08] where is the file you are trying to mv [08:08] if i go into the proccessed folder on the mount, and touch txt.txt it works [08:08] if i try to move a file from the root of the mount into processed mount root/proccessed i get operation not permitted [08:09] lotuspsychje can you repeat that at paul [08:10] what are you ralking about ptocessed mount [08:10] answet my question [08:10] wgete is the gule ypu are trying to mv [08:11] where is the file [08:12] i have a s3 bucket mounted using s3fs [08:12] ok [08:12] where is the file [08:12] /var/sftp/s3bucketname/ [08:12] the file is in that location [08:12] in /var/sftp/s3bucketname/ there is a folder called processed [08:12] ok [08:13] i can not move a file from /var/sftp/s3bucketname/ to var/sftp/s3bucketname/processed/ as I get operation ot permitted [08:13] I just created a new folder called test in /var/sftp/s3bucketname/ and tried to move a file to it, I can't do that either [08:13] how did you mount it [08:14] hey hey the cat [08:14] my imaginary cat moving imaginary files [08:15] gow did you mount the s3bucket [08:15] . /var/sftp/s3bucketmount fuse.s3fs _netdev,allow_other,multireq_max=5,iam_role=auto,complement_stat,uid=1500,gid=1500,umask=133 0 0 [08:16] that isnt root [08:16] but it worked, it's worked for a year till s3fs stopped working so i re mounted it [08:16] uid 1500 [08:16] that's from /etc/fstab [08:16] sorry it's mapped to the sftp user [08:17] it can be thr permissions of the amazon thing [08:18] i've checked with amazon and they said the s3 bucket setup is fine [08:18] it looks like you have some nonstandard amazon software to be mounting using special types [08:19] i'm using s3fs fuse to mount the s3 bucket [08:19] fuse.s3fs [08:19] correct [08:19] is that known by mount? [08:20] I dont know it [08:22] maybe the cat was listening to reach out and touch faith when it posed for the icecat logo [08:23] there are permissions on both sides [08:23] what user are you using to run mv [08:24] tried the sftp user and root [08:24] get same issue [08:25] and it is touchable, the file ? [08:26] if it is not moving with root that suggest it is the remote permissions [08:26] look into the permissions on the buckwt side [08:27] check the processed folder [08:28] the processed folder on the s3 bucket is fine [08:28] if i come outside the /var/sftp/s3bucketmount and run aws s3 cp test.txt s3://bucketname/ it works [08:28] it is owned by sftp [08:28] so it's something to do with that directory / chroot [08:29] no that isnt a valid test [08:29] you are attempting a remote move [08:29] not a local to remote copy [08:30] if i do local to remote from the s3bucket dir I get operation not permitted [08:30] I can touch a file in the root of s3 bucket [08:30] check the iamrole settings [08:30] I can go into processed and touch a file and it works [08:30] I just can not move from root of the s3 bucket into the folder in that same directory [08:30] whatever the s3fs manpage says [08:31] it has to do with remote permissions [08:31] what does the compaby do with this stuff [08:31] before I support it [08:32] I dont care to support moving porno files [08:33] paul if you read dune series you go deeper and deeper [08:34] exploring moral challenges in war scenarios [08:35] you come to the moral aspects of martyrdom and psychology mixed [08:35] very advanced subjects [08:36] -to be clothed in white [08:37] * soulseeker twitches [08:37] paul98 you are the company [08:39] amazon is not "real" but it is there [08:39] use it well === tinoco_ is now known as tinoco [12:17] teward: ping === stgraber_ is now known as stgraber [13:08] rbasak: cpaelzer: hi, I need to upload two no-change-rebuilds because of the clamav soname bump (somebody else uploaded new clamav, but it's a server package, so it being in migration for so long pings us) [13:09] how should I proceed with reviews? None? [13:09] they are c-icap-modules and pg-snakeoil [13:09] I just tested a local build, and they picked up the new libclamav9 dep just fine [13:09] ...and libclamunrar [13:20] Good question. [13:21] I'd say go ahead, no review needed. [13:21] k [13:22] rbasak: oh, one more [13:22] it's a deb sync [13:22] what will the rebuild version be like, just the usual -Nbuild1? [13:22] I think so, since there are no ubuntu changes [13:23] but I don't know debian's rule for no-change rebuids, if it's the same, we could collide [13:30] Debian doesn't do no-change rebuilds as they have binNMUs (unless something changed recently?) [13:30] -Nbuild1 is fine [13:30] "dch -R" should do the right thing [13:31] it did, thanks [13:39] People : hi ! Is that normal that the installer jumps back to the initial screen when one fills in the search domains when configuring manually the ip address of the machine ? And the same when trying to alter the partition table preset when selecting "full disk with lvm" ? [13:46] blackflow: ERR: No route to host. [13:47] blackflow: what's up? [13:53] geodb27: bug 1816954 - see comment 4 [13:53] bug 1816954 in subiquity (Ubuntu) "Unable to show Info nor Set static IP during install" [Undecided,Fix committed] https://launchpad.net/bugs/1816954 [13:53] (please don't cross-post) [13:54] geodb27: ^ yes, known issue where providing a search domain errors out. Set it without the search domain then set the search domain post-install [13:55] ^ this workaround won't work if you need to set a static ip address, though, use the one from comment 4 then [13:56] tomreyn: really? I've gotten subiquity to set the static IP if I give it the config WITHOUT a search domain [13:56] then set the search domain post-install [13:56] unless things've changed? [13:57] teward: oh i'm just going by what the error report discusses. [13:57] tomreyn: yeah, so if you set the *search domain* it will explode [13:57] "Also, when setting manual IPv4 information, selecting Save crashes the installer and starts over." [13:57] tomreyn: if and only if you put in the search domain [13:58] i see [13:58] tomreyn: so, in the 18.04.2 ISO's subiquity, static IP setting works *EXCEPT* if you put a search domain in [13:59] i understand that's what you're saying and i will believe it once i've tested it, too. ;-) [13:59] then it does some odd Python error I haven't been able to fully see because of the speed my VMs are efficient at (literally restarts before I get to dig into the error logs). [13:59] put in the static config without the search domain and it won't explode. [14:00] https://bugs.launchpad.net/subiquity/+bug/1816192 is probably the better bug [14:00] Launchpad bug 1816192 in subiquity "Installation crashes when setting the IPv4 settings to manual and having a value for search domains" [Undecided,Fix released] [14:00] but that isn't on the ISO(s) yet [14:03] so if you select "info" on the [14:04] ..NIC / ip address context menu it crashes, too [14:08] teward: ^ but you'Re right, it doesn't crash if you set a static ipv4 or ipv6 address, don't touch "info", don't set a search domain [14:08] right, that's the only thing I didn't test, but *setting* the IPv4 or v6 is doable [14:10] blackflow: responding to your ping, but i do have things to do today, if it's important email me via Launchpad or via my ubuntu.com address [14:55] morning! [14:55] anyone know how to run apache 1 in the foreground? [14:59] catbadger: (as previously discussed) there's no supported ubuntu version which comes with apache 1.x [15:00] you could try asking in #httpd, but i assume it's not supported there either. [15:01] lol hastag community [15:01] ... #httpd is another IRC channel :| [15:01] APache 1.x is **long since dead** though [15:02] so I wouldn't expect to find *anyone* able to assist you in this matter [15:02] already solved. [15:02] thanks anyway [15:03] thumbs will just send you away with fire once he finds out what you run ;) [15:03] * teward picks up the flamethrower [15:03] tomreyn: you assume I won't :P [15:03] i'd never dare to assume that! === sarnold_ is now known as sarnold [16:12] * sarnold notes that catbadger is now that httpd 1.x expert :) [16:13] hah [16:41] hi all [16:42] https://serverfault.com/questions/965608/trying-to-set-up-haproxy-letsencrypt-for-ssl-termination-gateway-not-answering i have a websrver running letsencrypt certs, I've set up a tcp proxy to allow the webserver to manage its own certs for now. in the interim, i'm setting up the proxy to terminate the ssl, this isn't working... i'm getting a dns error that makes no sense. http://leafdns.com/index.cgi?testid=A8086532 the actual [16:42] the error at certbot is 'Failed to establish a new connection: [Errno -3] Temporary failure in name resolution' [16:42] what does "The parent nameservers do not have your nameservers listed" mean? [16:44] can you ping the server u r trying to reach by it's name? number? none? generally, if you can ping numbers but not names, you need a dns (try 8.8.8.8) [16:44] catbadger: sounds like a dns messup. which domain is this? [16:44] catbadger: ^ [16:45] omg [16:46] it's my /etc/resolve.conf... why does it keep getting overwritten? [16:46] sepcifically the nameserver line [16:47] catbadger: it's usually controlled by whatever network config package you're using [16:47] so... netplan? [16:50] you can't edit that file - it says right in it "DO NOT EDIT THIS FILE BY HAND -- YOUR CHANGES WILL BE OVERWRITTEN" (and there is no "e" in resolve.conf. [16:50] catbadger: /etc/resolv.conf or /etc/resolve.conf? [16:50] i have nameservers: addresses [x.x.x.x,x.x.x.x] in my netplan config, but apply is writing nameserver 127.0.0.53 [16:50] resolv [16:50] what do you want to change it to and I can give you the command [16:50] catbadger: check out the systemd-resolved manpages [16:50] 8.8.8.8 8.8.4.4 [16:51] ok, you can do that with route command - I'll give an exact sample momentarily [16:51] thank you [16:54] soz - i don't have it, I put it in using thegraphical method under networks>configure network [16:55] i was thinking gateway not dns [16:56] To make the DNS related changes in resolv.conf permanent, you need to change the DHCP configuration file named dhclient.conf . You can find the file in /etc/dhcp/dhclient.conf . Open the file for editing (don't forget to use sudo). [16:56] thanks anyway [16:56] that's actually pretty helpful thanks [16:57] and this - but... i did it with the graphical interface and it worked fine on all the machines https://askubuntu.com/questions/130452/how-do-i-add-a-dns-server-via-resolv-conf [16:57] 8.8.8.8 and 8.8.4.4 [17:03] sigh [17:19] https://askubuntu.com/a/1001295/946188 has the ansewr [17:20] ty 4 sharing the answer [17:21] :) posterity etc. [17:23] i added it to my cheat sheet (speakin gof which - i recommend cht.sh shell to get working examples of things (over manpage) [17:59] what are the three parameters mean in preseed partitioning? [17:59] 40 50 100 ext3 [17:59] 40 M min, 100 M max .. and what's in the middle? and why need 3 parameters? [18:23] teward: I've asked in #nginx, when I pinged you, you can /lastlog me there, wanted to ask if you're aware of such a problem, this is Ubuntu Bionic [18:24] blackflow: /lastlog only works so long as it's in my logs, which #nginx is NOT stored in my ZNC logs for more than a bit. Your message wasn't in the scrollback [18:24] teward: tl;dr it appears that in a http2 request of a static html with many static png images, the images randomly error out with "timeout" error_logged and "200 0" access_logged, which I can't understand as it's all a single tcp connection [18:25] i'm not aware of any major HTTP/2 issues such as that [18:25] teward: I've since switched back http1.1 and there aren't any errors (but FF opens 5 connections to load up all the images on that page) -- the page is a munin status page with png graphs from rrd [18:26] > [2019-05-02 14:25:15] i'm not aware of any major HTTP/2 issues such as that [18:26] it SOUNDS like an issue with your client with the timeout errors being seen [18:26] but i've not had any reported issues such as that, and I'm pretty sure they'd have been bug-reported. You might want to try with the PPA version which is newer [18:26] I have to upload the latest mainline and stable to it but [18:26] teward: possible, but.... this was a FreeBSD server before it migrated to Ubuntu, same base version (1.14), and there were no such problems there with http2 [18:27] might have some bugfixes (use the mainline PPA) [18:27] blackflow: *I* have yet to encounter any such breakages, but without a munin setup myself to test with and a copy of your configs I can't give any real insights either way :| [18:27] that was my next step, to see if it's a bug fixed in newer version. then again, from what I know, http2 is still kinda experimental, and I've seen a few nginx bug in that area fixed, from the nginx mailinglist [18:27] I'm not *aware* of any such major problems (third time I've said this) which would cause the problem you're seeing. [18:28] blackflow: actually, HTTP/2 *itself* is considered stable, there's just OCCASIONAL things that pop up [18:28] it's the SRU rules that prohibit me from making nginx updates to newer upstream releases :P [18:28] teward: I'll replicate with an example I can share publically, this URL is behind firewall and auth, so I can't [18:29] blackflow: yep, a minimal reproducible example that would allow me to test, etc. would be great [18:29] because what if the problem is actually munin not liking it xD [18:29] teward: biggest problem is, it's not _every_ request. I thought it correlated to a packet loss I observed but then that makes no sense, that's tcp, lost packets would be retried. this is logged as client timeout, which is weird [18:30] teward: munin only creates static html+png. I've copied the page to exclude munin's cron overwriting png's mid-request, and it still appeared, so munin is out of the equation. [18:30] well, then provide a minimum reproducible example :P [18:30] so I can test/see if i can replicate [18:31] it's also entirely possible it *is* a client side issue on your side, since you haven't indicated what browser/OS you're connecting to it with [18:31] but as I said, i'm not aware of any major bugs in HTTP/2 in 1.14 that'd cause the problems you're experiencing [18:32] yup will do. Also, does this make any sense? run request + timeout within the same second? https://dpaste.de/zKzb [18:32] teward: Ubuntu Bionic on the client side too :) [18:32] (stock apt-installed Firefox) [18:33] blackflow: yes in some cases it can make sense [18:33] it's unusual but it usually means the client never sent ACK/SYN packets back (or they were lost in transit) in accepted timespansz [18:33] FF logged 17 second "receiving" before it gave up, in the dev console [18:33] but without wireshark/tshark logs on either side [18:33] hrm... that's why I thought it was this packet loss to blame but.... again.... being tcp, wouldn't lost packets be retried? [18:34] only so many times [18:34] you're thinking at a pure protocol level [18:34] there's only so many times packets are retried (time to live is a thing on TCP packets you know) [18:35] I'll prepare a public test page. [18:58] i have several overseas servers - they are unning postfix and mailutils. except for 1, they all work as expected. 1 always goes to junk in user's outlook, and cannot be junk unjunked. i printed and compared all the mail.cf files. The only thing unique on the 1 machine is "myhostname" does not end in .internal, but .local - could that be the real reason? [19:01] got it .. [20:39] I dont know think that would cause it. Why are they not ending in real TLD? [20:48] dax screwed things up again [20:49] a long list of messages sent to #ubuntu [20:49] is there any way to add rootdelay during lilo boot [20:49] after a fresh install it fails to boot not finding the partition [20:50] but by the time the boot fails the partition can br seen by the shell [20:50] paul98 are you there [20:50] prepare to exterminate mafia [20:51] all mafia everywhere [20:51] specificly watch police [20:51] they gravitate to positions of power [20:51] police and mayors [20:51] soulseeker: please rest your enter key. [20:54] tomreyn yeah ok, do you have any suggestionabout how to boot from lilo everything needed was explained [20:59] soulseeker: no, not personally, it's been years ago that i had to work with lilo [21:01] do you know hpw to continue the boot manuly mount returns invalid argument [21:02] it mounts with type specified but how to contnue the boot, what is the automated process doing [21:03] how to tell the initram to hand root to the mount [21:04] welcome back lucas gray [21:06] lucas_gray do you care to work with me for booting the fresh install [21:07] it halts booting and drops to shell because it does not find the boot partition it mounts fine manuly but what next how to contunue booting [21:13] all of these volunteers and nobody knpws how to continue booting [21:14] lucas_gray talk to me [21:14] It seems you are following me [21:14] you show up on various networks, talk to me [21:19] soulseeker: try in ##linux, too, if you can't find immediate support here. maybe there's also a lilo channel? [21:20] teward: ping? [21:21] lilo isnt the focus anymore it failed after booting I can install grub [21:21] lilo was selected because it detectsd sdb during indtall where grub was looking at sda [21:22] Id have used grub hut the installer had no option to select the right drive [21:22] if you read what Ive typed you can see where this is at sitting at initram shrll waiting for firstboot [21:23] what is the process to do it manuly [21:23] hi scotte whats going on [21:23] do you monitor this channel [21:24] things like this lucas_gray follow mr around from channel to channel buy do not respond to anything [21:25] there is nothing but boss attitudes making complaints lile this tomreyn [21:25] it is reminisciant of police [21:26] thrt claim to be protexting something but it isnt there [21:26] "keep this channel open for support" but no support [21:26] like americans protecting nonexistant freedom [21:27] scotte do you speak [21:28] hello scotte can you read [21:52] soulseeker: hopefully useful to you http://rurounijones.github.io/blog/2009/03/17/how-to-ask-for-help-on-irc/