Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2015-04-10

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

All times shown according to UTC.

Time Nick Message
00:01 rongze joined #fuel
00:34 teran joined #fuel
01:02 rongze joined #fuel
02:03 rongze joined #fuel
02:19 mattgriffin joined #fuel
02:27 mattgriffin joined #fuel
02:29 fedexo joined #fuel
02:36 Longgeek joined #fuel
02:49 mattgriffin joined #fuel
02:55 rongze joined #fuel
02:56 kun_huang joined #fuel
02:57 kun_huang which host os is fuel based on?
04:24 mattgriffin joined #fuel
05:19 fedexo joined #fuel
05:24 Longgeek joined #fuel
05:42 sambork joined #fuel
05:44 thumpba joined #fuel
05:47 gongysh joined #fuel
06:32 sc-rm evg: Found the problem to be that the switch was not tagging the l2 neutron 1000-1030 vlans
06:48 ibumarskov_ joined #fuel
06:48 bogdando joined #fuel
06:53 sambork joined #fuel
06:56 e0ne joined #fuel
07:00 kun_huang which host os is fuel based on?
07:04 e0ne joined #fuel
07:14 ibumarskov joined #fuel
07:15 geta-tn joined #fuel
07:15 geta-tn Hi
07:16 geta-tn evg: any good news ?
07:37 aliemieshko <kun_huang>hi, centos
07:37 kun_huang aliemieshko: got it, thanks
07:41 kevinbenton joined #fuel
07:43 teran joined #fuel
07:44 e0ne joined #fuel
07:55 sc-rm evg: Now I have another problem, which is that everytime I run the healt check the test uses an floating ip, but does not release it, so the start of the floating ip range keeps encreasing
07:58 e0ne joined #fuel
08:13 e0ne joined #fuel
08:18 Miouge joined #fuel
08:22 sc-rm evg: Also, the horizon is running quite slow even thought the controller nodes are quite fast, sas drives, 24G RAM and dual quadqores.
08:23 sc-rm evg: It’s slower that my old setup with pentium D, 8Gb RAM and sata drives
08:35 zhangqi joined #fuel
08:44 tzn joined #fuel
08:59 e0ne joined #fuel
09:02 Miouge joined #fuel
09:02 Miouge left #fuel
09:05 alecv joined #fuel
09:09 HeOS joined #fuel
09:21 stamak joined #fuel
09:22 geta-tn evg:are you there??
09:26 evg Hi
09:26 evg geta-tn: I've tried 5.1.1 as you told
09:26 geta-tn Hi
09:26 geta-tn and?
09:27 evg geta-tn: nothing like your issue
09:28 geta-tn what you mean?
09:28 evg geta-tn: resized all the disks in the env, deployed, then add a new compute with default partitioning, and deployed it
09:29 evg geta-tn: it didn't touch other nodes
09:31 geta-tn but if you reboot the node, fuel realizes that the controller has not any more the default config
09:31 geta-tn and when adding new node, asks to deploy changes in the disk config
09:32 evg geta-tn: what node I should reboot?
09:32 geta-tn the one you customize the disk config...
09:32 geta-tn for me was the controller
09:33 evg geta-tn: wait, I'll try
09:33 geta-tn can i send you here a snapshot??
09:34 evg geta-tn: yes, please
09:35 evg geta-tn: btw, do you use production or test environment?
09:35 geta-tn production
09:36 geta-tn how should i send the jpeg?
09:38 evg geta-tn: not sure, something like http://postimage.org/
09:41 geta-tn http://postimg.org/image/cg6kl0zbz/ http://postimg.org/image/vzb5udy3j/
09:41 geta-tn my customization was to use all the disk...not a part
09:42 geta-tn but now Fuel gives the default view and ask for deploying
09:53 evg geta-tn: fuel has to reinstall nodes to change partitioning. Something impossible.
09:53 evg geta-tn: all the controllers changed their configurations?
09:54 geta-tn yes
09:54 geta-tn we have 3 controllers and for all of them
09:55 geta-tn fuel wants to bring back the default config...
09:55 geta-tn it doesnt like the custom config
09:55 geta-tn :)
10:00 evg geta-tn: could you click on "sda" for more details
10:06 teran joined #fuel
10:08 evg sc-rm: Hi, sorry for my silence, don't get fo ignore. Really strange issue with geta-tn
10:09 evg sc-rm: what's the difference between new and old env? just versions?
10:09 teran_ joined #fuel
10:22 sc-rm evg: fuel 6.0 is the new version.
10:23 tobiash joined #fuel
10:24 sc-rm evg: What geta-tn is expericing is also what we have seen after a reboot of one of the controllers, fuel displays the default config of disk space even thought the controller was configured with a custome disk space.
10:28 Akshik joined #fuel
10:29 Akshik Hi, Im using Fuel 5.1
10:29 Akshik my external router gateway is showing down
10:30 Akshik is a HA environment,
10:31 evg sc-rm: thanks. I think the problem is fuel just show wrong info. It can't reinstall/redeploy nodes.
10:47 evg sc-rm: horizon could be slow because of keystone, its token's db swell.
10:47 evg sc-rm: http://docs.openstack.org/icehouse/install-guide/install/apt/content/keystone-install.html please see in the end
10:59 sc-rm evg: Hopefully it’s not the case, since the deployment only have been up for less than 2 days ;-)
10:59 sc-rm evg: And it has been slow since the begining
11:04 Miouge joined #fuel
11:19 e0ne joined #fuel
11:27 e0ne joined #fuel
11:42 vt102 joined #fuel
11:58 [HeOS] joined #fuel
12:12 rongze joined #fuel
12:18 dmitryme joined #fuel
12:38 sambork joined #fuel
13:03 evg sc-rm: tried healt check with floating ip, it was disassociated and deleted at the end. will try full set ot tests.
13:17 akupko joined #fuel
13:41 rongze joined #fuel
13:44 dmellado joined #fuel
13:59 sc-rm evg: Have tried a complete redeploy now and will do some more testing on monday, but I’ll leave for now.
14:02 teran joined #fuel
14:07 rongze joined #fuel
14:10 DaveJ__ joined #fuel
14:31 vidalinux joined #fuel
14:48 MiroslavAnashkin kybe: Yes. correct. Forgot about Juno. MOS major versions are always match with OpenStack major versions. 5.x - Icehouse, 6.x - Juno, 7.x - Kilo
14:59 teran joined #fuel
15:25 jobewan joined #fuel
16:13 zojoncj joined #fuel
16:42 daniel3_ joined #fuel
16:59 Akshik_ joined #fuel
17:12 Miouge joined #fuel
17:29 Akshik_ joined #fuel
17:40 e0ne joined #fuel
17:46 jskarpet joined #fuel
17:53 thumpba joined #fuel
17:57 e0ne joined #fuel
17:59 Akshik_ joined #fuel
18:29 Akshik_ joined #fuel
18:29 CheKoLyN joined #fuel
18:31 Guest58 joined #fuel
18:35 teran_ joined #fuel
18:49 Akshik__ joined #fuel
18:51 championofcyrodi If I have an HA controller that is using eth0 for the Public network, and would like to change it to eth1, how would I go about doing that?
18:51 championofcyrodi ovs-vsctl or something?
18:52 Akshik joined #fuel
18:55 championofcyrodi starting here: http://docs.mirantis.com/fuel/fuel-6.0/operations.html#how-slave-nodes-choose-the-interface-to-use-for-pxe-booting
19:03 championofcyrodi e.g. I've used pcs to move the vip__public resource to the server which has more bandwidth to support public connections.
19:03 championofcyrodi but the ethernet adapter it's using is 1Gbps, rather than the 10Gbps adapter i have connected to the same network switch.
19:03 championofcyrodi so i just want the network adapter the interface uses to change...
19:05 championofcyrodi also just noticed this: Clone Set: clone_p_neutron-l3-agent [p_neutron-l3-agent]
19:05 championofcyrodi p_neutron-l3-agent(ocf::mirantis:neutron-agent-l3):FAILED node-46.ostack.local (unmanaged)
19:05 championofcyrodi Started: [ node-50.ostack.local node-54.ostack.local ]
19:08 championofcyrodi looks like the "HA Proxy" is what i am referring to.
19:17 championofcyrodi astute.yaml seems to provide the information to cobbler... which implies the controller may need to be redeployed by cobbler... :(
19:27 HeOS joined #fuel
19:30 MiroslavAnashkin championofcyrodi: May be via /etc/udev/rules.d/70-persistent-net.rules on that controller?
19:32 championofcyrodi thats what i am looking at now
19:32 championofcyrodi but i'm scared it might break EVERYTHING!
19:33 championofcyrodi also what is interesting is the fuel UI shows the link as UP at 10.0 GBps when i click the Gear and expand the interfaces node...
19:33 championofcyrodi but when i actually click "Configure Interfaces" it shows the speed as N/A
19:33 Miouge joined #fuel
19:33 championofcyrodi as if both UI elements are populated from different backend services
19:34 championofcyrodi okay. o
19:34 championofcyrodi I'm DOING IT...
19:35 championofcyrodi should i just reboot after updating the udev rules?
19:35 MiroslavAnashkin Yes
19:35 championofcyrodi okay...
19:35 championofcyrodi here goes friday...
19:35 MiroslavAnashkin And you may be need to change cabling accordingly
19:44 e0ne joined #fuel
19:47 championofcyrodi already did the cabling... and it seems to be working...
19:47 championofcyrodi checking some network graphs now...
20:02 championofcyrodi so...
20:03 championofcyrodi my upload to containers via swift support through horizon showed abour 120.0MiB/s
20:03 championofcyrodi but when i download, its only around 30.0MiB/s
20:03 championofcyrodi (the same file)
20:03 championofcyrodi 1.5GB file
20:03 championofcyrodi so, essentially it uploads 3-4 times faster than downloads... why is that?
20:18 teran joined #fuel
21:23 HeOS joined #fuel
22:18 LiJiansheng joined #fuel

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