Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2014-09-26

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

All times shown according to UTC.

Time Nick Message
00:01 adanin joined #fuel
00:23 rmoe joined #fuel
00:58 klonhjcertusnet joined #fuel
01:59 jpf_ joined #fuel
02:00 adanin joined #fuel
02:48 jpf joined #fuel
02:53 jpf joined #fuel
02:56 harybahh joined #fuel
03:58 emagana joined #fuel
04:43 ArminderS joined #fuel
04:50 stamak joined #fuel
04:56 harybahh joined #fuel
05:15 AKirilochkin joined #fuel
05:16 anand_ts joined #fuel
05:36 stamak joined #fuel
06:19 dancn joined #fuel
06:28 emagana joined #fuel
06:31 sressot joined #fuel
06:36 e0ne joined #fuel
06:49 Longgeek joined #fuel
06:52 pal_bth joined #fuel
06:57 harybahh joined #fuel
06:59 flor3k joined #fuel
07:02 flor3k joined #fuel
07:02 AKirilochkin_ joined #fuel
07:05 warpc__ joined #fuel
07:13 pal_bth joined #fuel
07:18 sressot joined #fuel
07:44 harybahh joined #fuel
07:45 HeOS joined #fuel
08:22 stamak joined #fuel
08:27 AKirilochkin joined #fuel
08:28 f13o joined #fuel
08:29 emagana joined #fuel
08:35 Longgeek joined #fuel
08:37 adanin joined #fuel
09:02 e0ne joined #fuel
09:03 pal_bth joined #fuel
09:10 emagana joined #fuel
09:33 pal_bth joined #fuel
09:42 Longgeek joined #fuel
09:43 ddmitriev joined #fuel
09:53 t_dmitry joined #fuel
10:06 harybahh joined #fuel
10:10 emagana joined #fuel
10:17 Longgeek joined #fuel
10:49 flor3k joined #fuel
10:58 e0ne joined #fuel
11:00 teran joined #fuel
11:11 emagana joined #fuel
11:19 omelchek joined #fuel
11:29 Longgeek joined #fuel
11:59 pal_bth joined #fuel
12:12 emagana joined #fuel
12:26 stamak joined #fuel
12:37 flor3k joined #fuel
12:44 anand_ts hello all, I'm getting ntp synchronization error while installing fuel master node. Also it says more than dhcp servers detected and installation may fail. what could be the reason.
12:44 kaliya anand_ts: has the master access to the internet?
12:45 anand_ts I have 3 servers connected to a switch.
12:45 anand_ts kaliya: yeah
12:45 kaliya so from master you can ping the universe?
12:45 kaliya in case we should investigate in system logs
12:46 anand_ts kaliya: where I can see the installation logs?
12:48 kaliya anand_ts: should be in boot.log
12:49 anand_ts kaliya: okay. checking this. ty!
12:49 kaliya in messages, you should have more info about ntp and why is not working properly
12:50 kaliya you're always welcome
12:53 Dr_drache hmmm
12:53 kaliya Dr_drache: what are you hmmming for?
12:54 kaliya :)
12:54 Dr_drache no idea
12:54 Dr_drache I have no questions today :P
12:54 Dr_drache just thinking on anand_ts' questions.
12:55 merdoc hehe. day without local problems ((%
12:55 teran joined #fuel
12:55 teran_ joined #fuel
12:56 Dr_drache merdoc, i'm sure i'll come up with one
12:56 Dr_drache just for you
12:57 kaliya yes today we have a rather calm chat :)
12:57 kaliya maybe you guys are already in weekend mood?
12:58 Dr_drache kaliya, not at all
12:58 Dr_drache i'm running performance tests in instances now
12:58 kaliya Dr_drache: what are you using to run such tests
12:59 Dr_drache kaliya, straight SQL -> SAN tests, straight OS based. acual workloads.
13:00 Dr_drache I have known metrics from my KVM cluster, for the exact same workloads and qcow2 image.
13:00 Dr_drache nothing fancy or scripted.
13:13 emagana joined #fuel
13:13 anand_ts kaliya: pxe boot to slave nodes can I run from eth0 itself or it needs another interface?
13:13 Dr_drache anand_ts PXE boot is always the first interface.
13:14 anand_ts so only one interface required is it for configuring fuel
13:14 kaliya anand_ts: yes
13:15 merdoc Dr_drache: look at Rally. it's really nice https://wiki.openstack.org/wiki/Rally
13:16 anand_ts kaliya: one more doubt , there is static pool for node discovery and dhcp pool. my network admin gave only static IP's to configure. so how should I proceed?
13:17 Dr_drache anand_ts, if the PXE is on a segment with another DHCP, you might have a bad day.
13:18 kaliya anand_ts: you have to disable other DHCPs from where your PXE-booting hosts aer
13:18 anand_ts Dr_drache: yeah the same issue I guess, actually could you explain little bit on that. how to come out of this issue
13:18 kaliya anand_ts: there must be only the fuel master DHCP in function on that segment
13:19 kaliya anand_ts: are you working on metal or VMs?
13:19 anand_ts I just connected 3 servers to a switch and apart from master node I did network boot settings on other nodes
13:19 Dr_drache anand_ts, the easyest way to explain, is the the PXE/ADMIN/MANAGMENT networks either have to be their own VLANS, or seprate network. period
13:19 anand_ts kaliya: metal
13:19 Dr_drache anand_ts, you need vlans.
13:20 Dr_drache or, more switches.
13:20 kaliya anand_ts: Dr_drache is right, you must separate the networks. And on the segment where you're booting your servers, let the Fuel Master to release DHCP
13:21 Dr_drache anand_ts, the ips your network admin gave you, are only for the public network after fuel is setup, and maybe for a 2nd interface for fuel (which I have)
13:21 anand_ts kaliya: got it. Fuel master itself is acting as DHCP server here right
13:21 Dr_drache anand_ts, yes.
13:22 Dr_drache anand_ts, for a while I just used differnt switches in my testing... until I sorted out vlans.
13:23 Dr_drache I use GRE with tagged interfaces.
13:23 anand_ts Dr_drache: So can I use the same default IP which they gave, in that case how would I access the dashboard/
13:24 Dr_drache if you had a system on that network you could yes.
13:24 kaliya anand_ts: you can find some sample configurations for some switches here http://docs.mirantis.com/openstack/fuel/fuel-5.1/planning-guide.html#reference-configuration-of-hardware-switches
13:24 Dr_drache anand_ts, I setup a 2nd DHCP interface (eth1) for fuel, to access it from my live network.
13:26 HeOS joined #fuel
13:26 anand_ts Dr_drache: kaliya okay guys , thanks for the help. my network admin left for the day. I need to sit with him to clear up things, that is possible only on monday. So I thing its time to wind up and njoy weekend :)
13:28 Dr_drache awesome
13:28 kaliya anand_ts: let us know then
13:29 xarses joined #fuel
13:29 anand_ts kaliya: Dr_drache yeah sure. definitely need your help as well.
13:31 Dr_drache i'll try my besy
13:31 Dr_drache best
13:31 AKirilochkin joined #fuel
13:39 sc-rm When under high load, spawning new instances will faild with “no valid host could be found” even thought there are plenty of resources (storage, cpu, ram) but cephfs lack of response because of high load will cause that error...
13:49 emagana joined #fuel
13:59 adanin joined #fuel
14:01 HeOS joined #fuel
14:03 pal_bth joined #fuel
14:04 e0ne joined #fuel
14:10 e0ne joined #fuel
14:28 AKirilochkin joined #fuel
14:31 jaypipes joined #fuel
14:35 harybahh joined #fuel
14:36 jobewan joined #fuel
14:43 emagana joined #fuel
14:43 baboune left #fuel
14:44 mpetason joined #fuel
14:56 HeOS joined #fuel
15:00 emagana joined #fuel
15:07 xdeller_ joined #fuel
15:36 blahRus joined #fuel
15:45 Dr_drache hey guys
15:51 Dr_drache is there a way to mount NFS on the controller to migrate glace images?
15:51 Dr_drache glance
15:52 Dr_drache ahh, we do have some repos
15:58 rmoe joined #fuel
16:12 ArminderS joined #fuel
16:24 AKirilochkin joined #fuel
16:27 angdraug joined #fuel
16:34 AKirilochkin_ joined #fuel
16:36 harybahh joined #fuel
16:37 HeOS_ joined #fuel
17:06 Dr_drache mpetason
17:06 Dr_drache looks like the issue isn't fixed
17:07 merdoc Dr_drache: what issue do you have now? (%
17:08 Dr_drache merdoc, can't create any instances
17:08 merdoc O_O
17:08 Dr_drache "block device invalid"
17:08 merdoc smthg with storage backend?
17:09 Dr_drache but no errors from ceph or nova about it
17:09 Dr_drache I can create 60000000000000000000000000000000000000000000000000000000000000000000000000000 cirrOS instances
17:10 merdoc hm. I just realized that I never try to create win/linux instance after upgrading to 5.1...
17:16 merdoc s/try/tried/
17:25 teran joined #fuel
17:26 merdoc Dr_drache: win2012r2 started without any problems
17:29 Dr_drache merdoc, from what type of image?
17:31 merdoc raw
17:33 Dr_drache that could be the problem
17:33 Dr_drache these are qcow2s
17:34 merdoc ok. I have one qcow2 image
18:00 merdoc Dr_drache: you are right! custom qcow2 does not work.
18:00 Dr_drache it's not just me?
18:00 merdoc looking forward to monday digging (%
18:00 merdoc Dr_drache: yep
18:01 merdoc IOError: [Errno 28] No space left on device
18:01 Dr_drache I got 1 to work yesterday.
18:01 merdoc and that with 1.4Tb of free space on ceph (%
18:01 Dr_drache I was told to increase the size of the image (snapshot) drasticly
18:02 merdoc I have 8gb of qcow image and 40gb allocated in flavor
18:02 merdoc on 5.0.1 that image start on 20gb flavor with any problems
18:03 jpf joined #fuel
18:03 jpf joined #fuel
18:04 Dr_drache merdoc, good reference
18:04 Dr_drache merdoc, you have a way to convert that to a sparse raw?
18:06 merdoc what "sparse raw"? I have that image already converted into raw and they work. from that image I created first instance
18:09 Dr_drache sparse raw is like.... a raw qcow2
18:12 merdoc Dr_drache: dunno what you mean. I've create image as described here http://docs.openstack.org/image-guide/content/windows-image.html and then convert it to *.raw by qemu --convert
18:12 merdoc Dr_drache: some logs from compute node - http://paste.openstack.org/show/115671/ if you found someone who can deal with it
18:15 e0ne joined #fuel
18:17 Dr_drache merdoc, slightly differnt error log
18:17 Dr_drache but, same cause
18:20 e0ne joined #fuel
18:28 e0ne joined #fuel
18:36 harybahh joined #fuel
18:41 e0ne joined #fuel
18:49 AKirilochkin_ Guys, hi
18:50 Dr_drache AKirilochkin_ hey
18:50 e0ne joined #fuel
18:52 AKirilochkin_ I have a problem with installation, seems that this problem connected with big amount of drives.
18:55 AKirilochkin_ cobbler does not understand so long line in kickstart
18:55 AKirilochkin_ how i can fix this issue
18:56 AKirilochkin_ i am on-site with customer
18:56 Dr_drache ummm
18:58 AKirilochkin_ по-русски говоришь?
19:03 Dr_drache merdoc
19:03 Dr_drache my ceph is filled with unclean and stuck pages now
19:05 e0ne joined #fuel
19:11 ganso joined #fuel
19:15 e0ne joined #fuel
19:22 e0ne joined #fuel
19:26 boris-42 joined #fuel
19:31 e0ne joined #fuel
19:31 xarses joined #fuel
20:15 Dr_drache merdoc ; i'm not sure if this is related...
20:15 Dr_drache but I just had a network card die...
20:15 Dr_drache sooo
20:37 harybahh joined #fuel
21:13 kupo24z MiroslavAnashkin: is the commit https://bugs.launchpad.net/mos/+bug/1371723 avail in the latest community 5.1 stable ISO?
22:05 teran_ joined #fuel
22:37 harybahh joined #fuel

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