Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2014-06-02

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

All times shown according to UTC.

Time Nick Message
00:20 e0ne joined #fuel
01:18 e0ne joined #fuel
02:18 e0ne joined #fuel
02:48 e0ne joined #fuel
03:18 e0ne joined #fuel
03:46 IlyaE joined #fuel
04:18 e0ne joined #fuel
04:18 IlyaE joined #fuel
04:48 e0ne joined #fuel
06:56 e0ne joined #fuel
07:08 artem_panchenko joined #fuel
07:24 saibarspeis joined #fuel
07:37 izinovik joined #fuel
08:15 b-zone joined #fuel
08:18 izinovik joined #fuel
09:15 AndreyDanin joined #fuel
09:30 dnikishov joined #fuel
10:10 taj joined #fuel
10:35 e0ne joined #fuel
10:52 e0ne joined #fuel
11:58 e0ne_ joined #fuel
12:07 saibarspeis joined #fuel
12:58 vogelc joined #fuel
13:51 toonces joined #fuel
14:05 jaypipes joined #fuel
14:08 IlyaE joined #fuel
14:25 iunruh joined #fuel
14:26 iunruh hey all.. I'm on Fuel 3.x (can't upgrade) and I'm having an issue with log files not being rotated
14:26 iunruh from what I've gathered, rsyslog is aggregating the log files locally as well as shipping them off to the fuel master
14:27 iunruh I don't want to store files locally, does anyone know how to turn this functionality off?
14:28 jobewan joined #fuel
14:51 blahRus joined #fuel
14:52 evg iunruh: i'm afraid you have no way exept editing all rsyslog's conf files
14:57 e0ne joined #fuel
15:00 IlyaE joined #fuel
15:03 angdraug joined #fuel
15:05 e0ne joined #fuel
15:17 iunruh evg: I ended up just disabling rsyslog, since logs haven't appeared in Fuel web since like August of last year
15:30 crandquist joined #fuel
15:44 e0ne_ joined #fuel
16:10 IlyaE joined #fuel
16:25 IlyaE joined #fuel
16:30 rmoe joined #fuel
16:43 xdeller joined #fuel
17:05 IlyaE joined #fuel
17:09 mutex joined #fuel
17:12 angdraug joined #fuel
17:27 xarses joined #fuel
18:03 nullme joined #fuel
18:05 nullme Hello, is this support for Fuel 4.1 or you guys help out with 5.0 questions?
18:09 xarses nullme: This is the fuel channel, any version is fine =)
18:10 nullme I just tried to install 5.0 and looks to be good, but when i create my first cloud, it says "cloud has been setup but configuration has to be done manually" has anyone seen anything like this?
18:18 Kupo24z joined #fuel
18:20 xarses nullme: can you paste a screen shot of this?
18:21 Kupo24z Are there any possibilities of having Instance Snapshots being saved on a different Ceph storage pool from Glance Images?
18:22 e0ne joined #fuel
18:30 xarses Kupo24z: I'd gather that it can be possible, I just don't know how implemented it is. What is the use case?
18:34 IlyaE joined #fuel
18:38 Kupo24z xarses: We have an SSD storage pool and a slow sata disk storage pool, VM disks on the SSD with the glance images they spawn from and instance backups on the huge sata disks pool\
18:39 Kupo24z Id move glance to the sata pool but im afraid of performance issues with the VM's as they are thin provisioned and use the parent glance images for reads
18:40 xarses Kupo24z: and you're backing up the os volume ? I know cinder can do what you want already
18:41 Kupo24z OS is on the ephemeral disk, so cinder isnt used right now
18:41 xarses maybe boot from volume would work better for you
18:43 Kupo24z Wouldnt I have the same issues with Cinder? Or does it have support for booting volumes off one pool vs backing up volumes
18:43 Kupo24z off another*
18:47 xarses I was looking to see if glance supports multiple storage backends, and it looks like no. Cinder on the other hand does, so issue 1 that i see with glance is taken care of (#1, You need to be able to configure multiple storage drivers, even with RBD, you have to define each of the pools)
18:47 xarses #2 You need to be able to select the other pool when you make a snapshot
18:47 xarses cinder should also take care of this since they support backup's now
18:47 xarses but I havent played with that, only #1
18:50 Kupo24z xarses: would I lose the benefits of live migrations and thin provisioning with boot-from-volume then?
19:09 sanek joined #fuel
19:10 xarses for thin, I'm not sure, It would most likely matter for what you OSD structure looks like. If they are coming from the same OSD, IIRC then it will still thin. I haven't tested for certain but my observation implies that it does. for live migration, no you will still have that
19:26 nullme xarses: I can, Im uploading it and will send you link in 1 minute,
19:28 nullme xarses: https://5drive.5thcolumn.net/f/a8206122da/
19:31 xarses hrm, thats bad... Can you upload a diagnostic snapshot from the support page?
19:32 nullme yes
19:34 nullme https://5drive.5thcolumn.net/f/e8270cfb1b/
19:34 IlyaE joined #fuel
19:42 xarses nullme: what happens if you create another cluster with the same settings, and different name?
19:43 nullme let me try
19:43 nullme same thing, this cluser name was "test"
19:47 derrickb joined #fuel
19:48 IlyaE joined #fuel
19:53 xarses and from the command line if you do "fuel env create --name test2 --mode multinode --net neutron --nst vlan --rel 1"
19:57 nullme so it creates from command line, but with no configuration, same as from UI, it creates the basic but most of the info is missing
19:58 xarses what info is missing?
19:59 xarses can we try reseting the db?
19:59 xarses from the shell 'dockertcl shell nailgun'
19:59 xarses 'manage.py dropdb && manage.py syncdb && manage.py loaddefault'
19:59 nullme well i don't know if its missing, but the network screen in UI compared to 4.1 is different and missing field for floating IP's
20:00 nullme also if im in the UI under setting of cluster and then i click on networks tab, the screen goes blank... nothing shows up, i have to click on home and then cluster and networks..
20:01 xarses The floating IP should be just below the public network, or just near the bottom (depending on mode)
20:01 xarses lets try reloading the db then
20:02 nullme yes, floating IP's should be there but the whole item is missing..  ok lets reset db
20:05 sanek joined #fuel
20:06 nullme sorry don't know what you mean from 'docertcl shell nailgun'
20:07 xarses sorry, 'dockerctl shell nailgun'
20:08 nullme ok that worked, finished with done
20:09 xarses hard refresh your browser and try again
20:09 nullme ok, browser refreshed and same error
20:11 xarses What are you using as a browser?
20:12 nullme chrome
20:15 nullme WOW *** it was the browser,,, even after hard close and refresh, still had issues, tried IE and works fine...thanks for all you help...
20:17 xarses did you previously have a 4.1 cluster on the same host?
20:18 sanek joined #fuel
20:18 nullme not on the same host, but i did use 4.1 from the same browser, i have workstation connected to my lab with images and tools and that is what i was using for access
20:19 xarses odd, browser cache issues are usually related to reusing the same host identifier (ip/hostname)
20:21 nullme lesson learned, thank you again... now time to start testing 5.0   :)
20:21 nullme left #fuel
20:21 xarses thats odd that you had that much of a result, I switch between the release versions all the time
20:26 sanek joined #fuel
20:29 e0ne joined #fuel
20:38 sanek joined #fuel
20:44 sanek joined #fuel
20:55 sanek joined #fuel
21:40 e0ne joined #fuel
22:10 IlyaE joined #fuel
22:15 e0ne joined #fuel
22:20 xarses joined #fuel
22:22 xarses joined #fuel
23:00 e0ne joined #fuel
23:20 IlyaE joined #fuel
23:46 IlyaE joined #fuel
23:57 vogelc_ joined #fuel

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