Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster, 2015-01-09

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

All times shown according to UTC.

Time Nick Message
00:03 purpleidea joined #gluster
00:03 purpleidea joined #gluster
00:08 T3 joined #gluster
00:09 purpleidea joined #gluster
00:09 purpleidea joined #gluster
00:10 jvandewege joined #gluster
00:10 iPancreas joined #gluster
00:23 T0aD joined #gluster
00:33 iPancreas joined #gluster
00:58 jbrooks joined #gluster
01:01 purpleidea joined #gluster
01:07 purpleidea joined #gluster
01:07 purpleidea joined #gluster
01:07 iPancreas joined #gluster
01:12 bala joined #gluster
01:12 purpleidea joined #gluster
01:24 purpleidea joined #gluster
01:24 purpleidea joined #gluster
01:30 purpleidea joined #gluster
01:35 purpleidea joined #gluster
01:37 MattJ_EC joined #gluster
02:00 plarsen joined #gluster
02:04 nangthang joined #gluster
02:09 jobewan joined #gluster
02:18 harish joined #gluster
02:34 bharata-rao joined #gluster
02:35 haomaiwa_ joined #gluster
02:48 _Bryan_ joined #gluster
02:51 purpleidea joined #gluster
02:53 jbrooks joined #gluster
03:00 jackdpeterson Anyone have any luck with this sucker? E [nlm4.c:980:nlm4_establish_callback] 0-nfs-NLM: Unable to get NLM port of the client. Is the firewall running on client? OR Are RPC services running (rpcinfo -p)?
03:02 theron joined #gluster
03:10 bennyturns joined #gluster
03:19 eka joined #gluster
03:31 hagarth joined #gluster
03:40 RameshN joined #gluster
03:41 harish_ joined #gluster
03:48 suman_d joined #gluster
03:55 ppai joined #gluster
04:03 kanagaraj joined #gluster
04:05 hagarth joined #gluster
04:05 jobewan joined #gluster
04:07 jobewan joined #gluster
04:13 itisravi joined #gluster
04:13 atinmu joined #gluster
04:15 shubhendu joined #gluster
04:20 sakshi joined #gluster
04:22 gem_ joined #gluster
04:27 suman_d joined #gluster
04:32 soumya_ joined #gluster
04:32 rafi1 joined #gluster
04:34 jiffin joined #gluster
04:34 anoopcs joined #gluster
04:43 ndarshan joined #gluster
04:55 nishanth joined #gluster
04:56 meghanam joined #gluster
04:56 nshaikh joined #gluster
04:57 kumar joined #gluster
04:59 kalzz joined #gluster
05:01 prasanth_ joined #gluster
05:03 spandit joined #gluster
05:12 nangthang joined #gluster
05:21 kdhananjay joined #gluster
05:23 jaank joined #gluster
05:23 lalatenduM joined #gluster
05:25 Manikandan joined #gluster
05:41 nshaikh joined #gluster
05:43 nbalacha joined #gluster
05:59 kshlm joined #gluster
06:00 jobewan joined #gluster
06:10 dusmant joined #gluster
06:12 smohan joined #gluster
06:19 hagarth joined #gluster
06:26 ppai joined #gluster
06:30 vikumar joined #gluster
06:32 elico joined #gluster
06:32 glusterbot News from newglusterbugs: [Bug 1180394] [FEAT] Support for pNFS protocol in Glusterfs <https://bugzilla.redhat.com/show_bug.cgi?id=1180394>
06:37 rjoseph joined #gluster
06:38 suman_d joined #gluster
06:38 saurabh joined #gluster
06:42 deepakcs joined #gluster
06:47 atalur joined #gluster
06:48 overclk joined #gluster
06:54 dusmant joined #gluster
06:57 hagarth joined #gluster
06:59 kalzz joined #gluster
07:03 glusterbot News from newglusterbugs: [Bug 1180404] nfs server restarts when a snapshot is deactivated <https://bugzilla.redhat.com/show_bug.cgi?id=1180404>
07:04 anil joined #gluster
07:08 ctria joined #gluster
07:14 soumya_ joined #gluster
07:17 meghanam joined #gluster
07:18 vikumar joined #gluster
07:21 eryc joined #gluster
07:21 eryc joined #gluster
07:22 jtux joined #gluster
07:32 dusmant joined #gluster
07:33 glusterbot News from newglusterbugs: [Bug 1180411] CIFS:[USS]: glusterfsd OOM killed when 255 snapshots were browsed at CIFS mount and Control+C is issued <https://bugzilla.redhat.com/show_bug.cgi?id=1180411>
07:33 bala joined #gluster
07:34 rgustafs joined #gluster
07:44 SOLDIERz joined #gluster
07:45 SOLDIERz hey everyone got at the moment a really odd problem with glusterfs and don't know how to fix it actually I got a 12 node gluster where to nodes reporting a splitbrain for two file dircetories
07:45 SOLDIERz where the other seems healty
07:45 SOLDIERz How I'm able to fix it
07:46 SOLDIERz just delete the files from the mount or directly in brick-volume .glusterfs
07:54 [Enrico] joined #gluster
07:54 Philambdo joined #gluster
07:54 [Enrico] joined #gluster
08:01 ppai joined #gluster
08:09 lkoranda joined #gluster
08:09 meghanam joined #gluster
08:11 soumya_ joined #gluster
08:18 deniszh joined #gluster
08:23 soumya_ joined #gluster
08:32 rgustafs joined #gluster
08:33 glusterbot News from newglusterbugs: [Bug 1180424] Glusterd-mismatch between arguments and parametrs in some functions (ex:glusterfs_uuid_buf_get,glusterfs_lkowner_buf_get). <https://bugzilla.redhat.com/show_bug.cgi?id=1180424>
08:39 T3 joined #gluster
08:42 fsimonce joined #gluster
08:42 badone joined #gluster
08:44 T0aD joined #gluster
08:49 Anuradha joined #gluster
08:53 Slashman joined #gluster
08:55 spandit joined #gluster
09:05 ppai joined #gluster
09:06 Norky joined #gluster
09:10 georgeh joined #gluster
09:12 aravindavk joined #gluster
09:21 calum_ joined #gluster
09:30 anti[Enrico] joined #gluster
09:35 karnan joined #gluster
09:44 hagarth joined #gluster
09:46 dusmant joined #gluster
09:51 nishanth joined #gluster
09:52 bala joined #gluster
09:57 badone joined #gluster
10:01 lalatenduM joined #gluster
10:02 nshaikh joined #gluster
10:14 Marqin joined #gluster
10:14 Marqin Hello
10:14 glusterbot Marqin: Despite the fact that friendly greetings are nice, please ask your question. Carefully identify your problem in such a way that when a volunteer has a few minutes, they can offer you a potential solution. These are volunteers, so be patient. Answers may come in a few minutes, or may take hours. If you're still in the channel, someone will eventually offer an answer.
10:15 deepakcs joined #gluster
10:16 Marqin I got lots of http://pastebin.com/AuLfsvWR in my logs and many D-state processes and when I try to rm that file rm goes into D, but I can Read and copy that file.
10:16 glusterbot Please use http://fpaste.org or http://paste.ubuntu.com/ . pb has too many ads. Say @paste in channel for info about paste utils.
10:16 Marqin argh
10:16 Marqin http://pastebin.com/raw.php?i=AuLfsvWR
10:16 glusterbot Please use http://fpaste.org or http://paste.ubuntu.com/ . pb has too many ads. Say @paste in channel for info about paste utils.
10:16 Marqin glusterbot: on raw are no ads
10:29 Marqin ok, after reboot it's working now
10:29 Marqin but what happened
10:47 ctria joined #gluster
10:53 nrcpts joined #gluster
10:55 soumya joined #gluster
10:58 MattJ_EC joined #gluster
10:59 dusmant joined #gluster
10:59 nishanth joined #gluster
11:00 bala joined #gluster
11:06 rgustafs joined #gluster
11:11 gothos btw. is there any info on 3.6.2? I need a good justification for my boss to restart the gluster system :D
11:16 ctria joined #gluster
11:25 smohan_ joined #gluster
11:38 hagarth gothos: 3.6.2 will be out next week
11:39 hagarth that is the tentative plan as of now :D
11:45 soumya_ joined #gluster
11:56 SOLDIERz joined #gluster
12:20 dusmant joined #gluster
12:24 bala joined #gluster
12:34 suman_d joined #gluster
12:37 calisto joined #gluster
12:38 itisravi joined #gluster
12:39 kanagaraj joined #gluster
12:40 RameshN joined #gluster
12:41 T3 joined #gluster
12:42 LebedevRI joined #gluster
12:43 jmarley joined #gluster
12:51 chirino joined #gluster
12:55 Anuradha joined #gluster
13:00 rjoseph joined #gluster
13:01 anoopcs joined #gluster
13:06 John_HPC joined #gluster
13:15 ppai joined #gluster
13:23 anoopcs joined #gluster
13:25 jmarley joined #gluster
13:28 harish_ joined #gluster
13:30 plarsen joined #gluster
13:35 aravindavk joined #gluster
13:46 primusinterpares joined #gluster
13:51 [Enrico] joined #gluster
13:56 nshaikh joined #gluster
14:07 redgoo joined #gluster
14:08 redgoo I need help " I have a replicated gluster volume with 3 peers but now we dont need the 3rd peer so would like to drop it out, if i use the remove_brick command or drop peer would that cause any issues for me and my content??? "
14:08 redgoo any help is much appreciated
14:16 bene joined #gluster
14:26 hchiramm joined #gluster
14:37 deniszh left #gluster
14:39 theron joined #gluster
14:39 giddy joined #gluster
14:41 giddy hellp
14:41 giddy hello
14:41 glusterbot giddy: Despite the fact that friendly greetings are nice, please ask your question. Carefully identify your problem in such a way that when a volunteer has a few minutes, they can offer you a potential solution. These are volunteers, so be patient. Answers may come in a few minutes, or may take hours. If you're still in the channel, someone will eventually offer an answer.
14:42 giddy when mounting gluster with glusterclient via /etc/fstab can you include more than just primary server + backup server?
14:42 jmarley joined #gluster
14:53 pcaruana|afk| joined #gluster
14:56 dgandhi joined #gluster
14:57 haomaiwang joined #gluster
15:10 shubhendu joined #gluster
15:13 prg3 joined #gluster
15:16 lpabon joined #gluster
15:16 haomaiwa_ joined #gluster
15:21 haomaiwa_ joined #gluster
15:21 lalatenduM joined #gluster
15:24 jobewan joined #gluster
15:35 aravindavk joined #gluster
15:41 nbalacha joined #gluster
15:43 tdasilva joined #gluster
15:44 DougBishop joined #gluster
15:47 lmickh joined #gluster
15:47 nrcpts joined #gluster
15:48 jmarley joined #gluster
15:48 nrcpts joined #gluster
15:49 virusuy joined #gluster
15:49 virusuy joined #gluster
15:51 sputnik13 joined #gluster
15:54 hagarth joined #gluster
16:13 T3 joined #gluster
16:14 bala joined #gluster
16:23 deniszh joined #gluster
16:30 RameshN joined #gluster
16:33 _Bryan_ joined #gluster
16:42 Arminder joined #gluster
16:43 mrEriksson joined #gluster
16:49 partner giddy: i would strongly suggest you to use DNS to provide names of all the boxes you have
16:49 semiosis partner++
16:49 glusterbot semiosis: partner's karma is now 1
16:50 partner wow, thanks, only took couple of years to reach this level \o/ :D
16:53 partner giddy: ie. round-robin entry to dns that is what you configure to your clients fstab and the record returns any/every gluster server you have on the peer serving the volume
16:57 deniszh left #gluster
16:58 giddy thank you partner
16:58 deniszh joined #gluster
16:58 giddy could anybody point me to a recent guide on round robin dns + glusterfs ?
16:58 giddy partner++
16:59 glusterbot giddy: partner's karma is now 2
17:06 partner giddy: basically you just create A records with some common name with ip addresses pointing to your gluster servers
17:06 theron joined #gluster
17:07 partner for example like this (random google hit): http://www.zytrax.com/books/dns/ch9/rr.html#services
17:08 partner they all hold the same information about all the volumes in the peer so reaching any is enough
17:11 Telsin you think that's better than using something ctdb or keepalived to provide a virtual IP for entry? (certainly easier)
17:12 partner definately, its not NFS
17:13 partner all the gluster servers know everything there is to know about so reaching any is enough
17:13 partner its a different case if one is not using native mounts ie. for example NFS
17:13 Telsin ah, for gluster mounts, gotcha. that makes sense. I'm still supporting a mix, so I guess I keep my keepalived a bit longer
17:14 partner even with NFS (if using glusterfs provided NFS) its enough to reach any of the servers and it will deliver the real mount point via that server (thought that costs some extra networking)
17:14 partner but no fault-tolerancy there "as-is"
17:15 Telsin yeah, that's the catch. with ovirt, it's proved happier to have the fault tolerance for me
17:19 giddy thanks partner
17:20 deniszh joined #gluster
17:31 marbu joined #gluster
17:36 lkoranda joined #gluster
17:37 mbukatov joined #gluster
17:38 shubhendu joined #gluster
17:43 semiosis for future reference, ,,(rrdns)
17:43 glusterbot You can use rrdns to allow failover for mounting your volume. See Joe's tutorial: http://goo.gl/ktI6p
17:47 gothos one might believe that gluster is already doing that internally due to the volume knowing about the different gluster servers
17:50 semiosis but how does the client know about servers before it fetches the volume info?
17:53 gothos sure, for such situation it's hands down useful
17:53 sputnik13 joined #gluster
17:58 nrcpts joined #gluster
17:59 plarsen joined #gluster
18:00 coredump joined #gluster
18:10 nueces joined #gluster
18:13 ekuric joined #gluster
18:26 Gill joined #gluster
18:27 eka joined #gluster
18:27 Gill Hey guys! I’m having a really frustarting issue. I am trying to mount a glusterfs client on a linode in NJ to a glusterfs server in AZ. there is 60 ms latency and I have an openVPN. I keep getting the error connection timed out though. Can anyone please point me in the right direction? thanks so much!
18:27 Gill these are the errors:
18:27 Gill [2015-01-09 18:27:21.280517] E [socket.c:1685:socket_connect_finish] 0-glusterfs: connection to  failed (Connection timed out)
18:27 Gill [2015-01-09 18:27:21.280688] E [glusterfsd-mgmt.c:740:mgmt_rpc_notify] 0-glusterfsd-mgmt: failed to connect with remote-host: Transport endpoint is not connected
18:48 partner not the expert on that area, have you tried playing with timeout values?
18:49 jmarley joined #gluster
18:54 Gill hi partner I have been looking for the timeout valiues cant find any that are specifically for mounting
18:55 ricky-ticky joined #gluster
19:05 JoeJulian Gill: Can you ping all of the servers from the client? Can you telnet to port 24007 from the client to the servers?
19:05 Gill JoeJulian:  yes i can ping
19:05 glusterbot Gill: Please don't naked ping. http://blogs.gnome.org/markmc/2014/02/20/naked-pings/
19:06 JoeJulian heh, regex fail.
19:06 Gill i got 60 ms latency on one side and 117ms latency to the other
19:06 Gill oht hat link wasnt for me?
19:07 Gill trying to telnet now
19:07 Gill so far nothing
19:07 Gill i have other clients that are more local to the servers that are connected
19:08 Gill this is the 3rd client im trying to bring up
19:08 ricky-ticky2 joined #gluster
19:08 Gill telnet doesnt seem to be working
19:09 Gill telnet HOST 24007 right?
19:10 Gill connectin timed out
19:11 Gill locally i can telnet to it
19:13 theron joined #gluster
19:13 Gill wow its been a long week :( firewall rules!!!
19:14 JoeJulian :)
19:15 Gill thanks so much JoeJulian!! says its connected now… doing a df -h now its takes forever though
19:16 Gill it wont let me to go to the mount point
19:16 Gill is that normal?
19:16 Gill 0-fs_storage-client-1: connection to  failed (Connection timed out)
19:16 Gill :?(
19:16 JoeJulian more firewall
19:17 lkthomas joined #gluster
19:17 Gill oh wait
19:17 Gill i just didnt wait long enough
19:17 Gill it connected then times out then came back online
19:17 Gill is that normal behavior?
19:18 M28 joined #gluster
19:18 JoeJulian It was waiting for the connection to that server. It timed out so it just worked with the remaining one.
19:18 JoeJulian The client connects to all the servers in the volume.
19:19 Gill Oh
19:19 Gill so that server is only connected to one of the servers now?
19:19 JoeJulian client is only connected to one server, yes.
19:20 JoeJulian as far as I can tell
19:20 JoeJulian but the symptoms fit.
19:21 Gill ah ill try a restart and put the mount in fstab and try again
19:21 Gill thank you so much! I;ve been fighting it for 5 hours at least
19:21 Gill i feel so stupid
19:21 Gill but greateful!
19:21 JoeJulian Happens all the time.
19:21 JoeJulian You're in very good company.
19:22 Gill thanks
19:22 Gill is bg a good idea to have int he options
19:22 Gill i read that it will continue to try to mount the volume in the background ( it needs to mount through VPN)
19:23 JoeJulian I'm not sure. That sounds distro specific.
19:24 Gill ah ill try without to start
19:30 coredump|br joined #gluster
19:30 PatNarciso Happy Friday all.   Does anyone know if there is a detailed setup doc related to "Cutting Edge"'s setup?  [ref: http://blog.gluster.org/2013/12/glusterfs-keeps-vfx-studio-on-the-cutting-edge/]
19:31 Gill JoeJulian: this VM is still starting… [ ok ] Starting haproxy: haproxy.
19:31 Gill Starting nginx: nginx.
19:31 Gill [ ok ] Starting NTP server: ntpd.
19:31 Gill [....] Mounting nfs01.az.internal:/fs_storage at  /mnt/nfs/storage/...[....] Sta[ ok  deferred execution scheduler: atd  guess its waiting to mount?
19:32 plarsen joined #gluster
19:33 JoeJulian PatNarciso: I haven't seen one. Sounds like they just installed standard distribute volumes at both locations though.
19:34 JoeJulian Gill: probably... Are you mounting nfs, or did that just happen to be the hostname?
19:34 Gill JoeJulian:  the hostname is nfs01.az.internal
19:35 Gill but the server needs to connect to the VPN first
19:35 JoeJulian Ah, I see.
19:35 PatNarciso JoeJulian: right.  I'm curious if they did anything special related to performance.   nufa, write-behind... that sorta thing.
19:35 Gill maybe bg would of been a good option lol let me go into rescue mode to change it up
19:35 JoeJulian PatNarciso: they didn't say.
19:37 JoeJulian A lot of companies are happy to say, "We used this open source software to solve a problem!" but frequently hide the details since their engineering time, they feel, gives them a competitive edge.
19:37 JoeJulian ... a frustration I have with my current employer.
19:38 Gill JoeJulian:  sorry for all the qestions but should i have a backupvolfile-server set too?
19:38 Gill or will glusterfs just automatically connect to both?
19:38 PatNarciso I feel like glusterfs is something I need to implement asap; and am happy to document the absolute crap out of it.
19:38 JoeJulian I prefer the ,,(rrdns) method, myself, but without that you would want the backupvolfile-server.
19:38 glusterbot You can use rrdns to allow failover for mounting your volume. See Joe's tutorial: http://goo.gl/ktI6p
19:39 Gill awesome i should be able to do rrdns
19:39 Gill oh wow its just an A record for each? no NAPTR or DNS SRV stuff?
19:40 Gill AMAZING!!!lol
19:40 JoeJulian :)
19:40 Gill JoeJulian:  youre making my friday
19:41 Gill i think its OK
19:42 PatNarciso hmm.  is it possible with nufa to have all writes go to an SSD-brick; and then (perhaps during rebalance) have the SSD-brick emptied out? (v 3.4.2)
19:42 PatNarciso [noteworthy: i'm staying away from xfs]
19:42 JoeJulian nufa is deprecated.
19:43 PatNarciso awesome.
19:43 JoeJulian Why are you staying away from xfs?
19:43 JoeJulian There's a new tiered storage effort for 3.6
19:44 JoeJulian er, 3.7
19:44 JoeJulian http://www.gluster.org/community/documentation/index.php/Planning37
19:44 PatNarciso why away from xfs?  cash money; we can fit 4 hhd's into an inexpensive node.  xfs would up the cost a bit, using 2 of the sata's for the performance gain (if I understand xfs correctly)
19:45 PatNarciso a sticky note of: april 29th, gluster 3.7.0 GA!   is on the wall of my office.
19:45 Gill JoeJulian: bg aint working either
19:45 Gill this is weird
19:46 JoeJulian Perhaps don't auto-mount and instead mount in rc.local?
19:47 Gill JoeJulian:  how would i do that?
19:47 JoeJulian PatNarciso: xfs is just a filesystem. Yes, it has the ability to put the journal on another device, but I've never heard of anyone actually doing that.
19:47 ildefonso joined #gluster
19:48 JoeJulian Gill: Add the "noauto" mount option in fstab, then just put the "mount $gills_mountpoint" in rc.local (assuming the vpn is supposed to be up by then)
19:50 Gill cool trying to figure out when openvpn will connect now
19:51 PatNarciso I said xfs, I was thinking zfs.
19:51 JoeJulian Oooooh, ok.
19:56 neofob left #gluster
20:00 javi404 joined #gluster
20:00 javi404 hello my gluster friends
20:01 semiosis hi
20:01 glusterbot semiosis: Despite the fact that friendly greetings are nice, please ask your question. Carefully identify your problem in such a way that when a volunteer has a few minutes, they can offer you a potential solution. These are volunteers, so be patient. Answers may come in a few minutes, or may take hours. If you're still in the channel, someone will eventually offer an answer.
20:01 semiosis javi404: ^^^
20:01 javi404 how can I force remove a dead peer (we won't be replacing it) do i have to resize it first? its currently 1 x 6 = 6.
20:02 semiosis remove-brick force?
20:02 Gill JoeJulian: I am changing the init scripts to make openvpn start sooner
20:03 javi404 lets ee
20:03 semiosis javi404: https://github.com/gluster/glusterfs/blob/master/doc/admin-guide/en-US/markdown/admin_managing_volumes.md#shrinking-volumes
20:03 javi404 semiosis: thanks
20:04 semiosis yw
20:04 Gill no dice.. gonna use your method JoeJulian hopefully the openvpn is up by then
20:05 PatNarciso not related to my situation above, just curious: what do you favor-- data redundancy via node, or data redundancy via raid.
20:05 glusterbot PatNarciso: favor's karma is now -1
20:06 PatNarciso glusterbot, good.  thanks.
20:06 semiosis Gill: perhaps you want to hook into openvpn with --route-up & --route-pre-down
20:06 JoeJulian Depends on the use case.
20:06 JoeJulian Oooh, nice idea.
20:07 Gill semiosis: how would I do that?
20:07 Gill this system should basically always be on the OpenVPN
20:07 semiosis Gill: consult openvpn man page
20:07 Gill cool
20:07 semiosis there's more than one way to trigger something to happen when the tunnel goes up or down
20:07 semiosis i've never done it myself, though, so i idk which is right for you
20:08 PatNarciso true.
20:08 Gill looks pretty cool
20:08 semiosis oh, here we go, jump to the "SCRIPTING AND ENVIRONMENTAL VARIABLES" section of https://community.openvpn.net/openvpn/wiki/Openvpn23ManPage
20:09 semiosis Gill: ^
20:10 semiosis looks like i was right the first time with route-up & route-pre-down
20:10 PatNarciso so if nufa is depreciated... hmm... is there a middle ground before 3.7.0.
20:11 JoeJulian nope
20:12 PatNarciso now thats my kinda development.
20:13 JoeJulian nufa was abandoned because it wasn't being used and wasn't reliable. Efforts were focused on the features being used.
20:13 JoeJulian As is always the case with open-source, feel free to contribute. :P
20:13 Gill so semiosis I can run mount -a as a route-up comd right?
20:13 PatNarciso of course; makes sense.
20:14 Gill and add noauto
20:14 semiosis Gill: if you're asking for my permission, you have it.  otherwise i really dont know :)
20:14 Gill haha thanks. Just making sure I understood what you were getting at with the route options
20:15 semiosis yes, thats the idea, or maybe make a script with your mount command
20:17 Gill cool ill give it a try
20:17 sputnik13 joined #gluster
20:26 DV joined #gluster
20:33 theron joined #gluster
20:36 PatNarciso is ALU is still gonna be around in 3.7?
20:38 * PatNarciso needs to stop reading docs from 2009
21:03 semiosis JoeJulian, anyone: (OT) recommendations for a user-friendly open source router firmware/distro?
21:06 JoeJulian I've been using the vyatta clone on the edgerouter lites. Seems pretty functional at its base, and is linux so any additional features I want I can have.
21:07 semiosis JoeJulian: link?
21:08 JoeJulian It comes installed...
21:08 semiosis oh
21:09 semiosis JoeJulian: does it do openvpn TUN?
21:09 semiosis istr one of these things that only did TAP or TCP or some such useless thing.  i need TUN over uDP
21:10 JoeJulian oh, wow... the core looks pretty dead.
21:10 semiosis https://blog.dlasley.net/2013/06/configure-openvpn-with-x-509-ubiquiti-edgerouter-lite/ looks promising
21:10 semiosis thanks for the tip i'll consider one of these
21:12 MattJ_EC joined #gluster
21:13 JoeJulian I love 'em. I used a bunch a Ed Wyse and I replaced my aging Cisco with one here at home.
21:18 Gill thanks again for the help semiosis and JoeJulian !
21:18 semiosis yw
21:18 Gill have a great weekend!
21:22 purpleidea semiosis: just curious, why specifically TUN over UDP? (fyi: socat)
21:23 semiosis pretty sure TUN over UDP is the way most people use openvpn, but i came across one of these ubiquity devices (iirc) that only did TCP, which is insane.
21:23 purpleidea that is insane
21:23 semiosis http://sites.inka.de/bigred/devel/tcp-tcp.html
21:24 JoeJulian I typically use strongswan for ipsec vpns.
21:24 semiosis incidentally, my last employer insisted that we use tcp for openvpn. one of the reasons i quit :)
21:24 purpleidea yeah, generally i use the TCP option when UDP is blocked
21:24 purpleidea JoeJulian: i've met the strong/libre/free swan dude, he's canadian. he works for redhat
21:24 semiosis someone suggested running my ovpn server on UDP/53 to avoid blocks :)
21:24 semiosis but i've never been blocked, airports, hotels, schools, etc
21:24 semiosis JoeJulian: i dont like ipsec
21:25 semiosis but yeah when i have to i use a swan
21:25 semiosis been a while, i dont remember which
21:25 purpleidea semiosis: you might want to read: https://nohats.ca/wordpress/blog/2014/12/29/dont-stop-using-ipsec-just-yet/
21:26 purpleidea (written by that canadian swan dude)
21:28 semiosis heh
21:30 semiosis ever used pfsense?
21:30 jackdpeterson joined #gluster
21:31 semiosis bsd based fw distro
21:37 purpleidea it's junk :P bsd is deprecated, i thought, no?
21:37 purpleidea there are some really cool things there, but it's sadly not linux
21:37 JoeJulian I haven't used it, but I'm reading a lot of people love it.
21:37 purpleidea i don't think it beats nftables+shorewall+puppet-shorewall
21:38 purpleidea but i do agree iptables was always a bit of a pita to use, hence the shorewall usage ^^
21:39 semiosis 1
21:39 semiosis wrong window
21:40 semiosis purpleidea: beats?  by what measure?
21:40 purpleidea semiosis: arbitrary purpleidea units
21:40 semiosis ok right
21:41 semiosis i really just want something with a friendly UI that can do normal openvpn stuff.  my beloved openwrt dropped openvpn from the webui, so i'm moving on
21:41 purpleidea semiosis: the serious answer is that if i had machines that *only* did pfsense and nothing else, then it might be a close call (still a loss for bsd though) but since there are typically other things on the machines, then those things work better on GNU/Linux
21:42 purpleidea make sense?
21:42 semiosis sure
21:42 purpleidea semiosis: tomato?
21:43 semiosis purpleidea: latest stable release is 4 years old according ot wikipedia :(
21:43 semiosis and it's got proprietary code like dd-wrt, so bleh
21:44 purpleidea semiosis: didn't know either of things were true! yikes! i guess it's time to switch :(
21:44 semiosis lol
21:46 purpleidea i actually use a GNU/Linux box as my personal home internet facing firewall, and then after my traffic passes through there, i have the tomato for wireless and easy home nonsense crap
21:46 purpleidea that GNU/Linux box does irssi, torrenting of CentOS iso's, etc, so it doesn't have to pass through the crappy router
21:46 jmarley joined #gluster
21:47 semiosis JoeJulian: know if theres any way to play with the edgerouter os without their hardware?  in a vm?
21:49 JoeJulian I think qemu does mips64
21:50 JoeJulian Oh, but it's going to pull stuff from cmos, so probably not.
21:51 JoeJulian But hell, they're cheap. Order one from amazon and send it back if you don't like it.
21:51 semiosis probably will
21:52 PatNarciso ehh? watcha buyin?
21:52 semiosis looking to upgrade my office firewall.  need easy webui & openvpn support.  open source.
21:53 semiosis JoeJulian recommends edgerouter lite, and i'm currently evaluating pfsense
21:54 PatNarciso edgerouter lite looks bad ass.
21:54 purpleidea semiosis: i haven't used the ubnt stuff myself, but a pal of mine does and is very pleased with it. i'd use it over bsd
21:55 semiosis purpleidea: heh, i've been running openbsd firewalls for years so i'm partial to the bsds
21:55 PatNarciso hmm... only I would see 99.9% uptime as a bad thing.
21:55 semiosis yep, linux uptime measured in days, bsd uptime measured in years :)
21:57 JoeJulian I have ubnt APs that have been up for 5 years.
21:57 purpleidea 17:02:45 up 447 days, 38 min, 10 users,  load average: 0.14, 0.13, 0.10
21:57 purpleidea i guess it's time for a reboot
21:57 semiosis lol
21:58 PatNarciso shit no.
21:58 purpleidea and an upgrade off of a P4
21:58 purpleidea but it's GNU/Linux
21:58 JoeJulian Looking forward to rolling kernel updates so we never have to reboot again.
21:58 purpleidea +1
21:58 purpleidea JoeJulian: but how will you upgrade glibc?
21:59 JoeJulian The same way? I don't know. That's for the devs to copy from the kernel.
21:59 purpleidea i don't know either, but i didn't think it was wholly possible yet
22:00 JoeJulian It's a bunch on 1s and 0s. How hard can it be?
22:00 purpleidea but i think the future future is really nothing to do with rebootless kernel upgrades, but rather a cluster of machines that host vm's or containers or both, and when you want to upgrade your base OS, you move the apps off, and then upgrade/reboot, and then load balance them back on
22:01 purpleidea eg: project atomic
22:02 JoeJulian You're not going to drag me kicking and screaming into lxc.
22:02 purpleidea i won't drag you ever!
22:02 purpleidea hehe
22:03 JoeJulian I've toyed with using docker for some of my home services and it just flat out failed.
22:03 JoeJulian Just not satisfied with network configuration.
22:03 purpleidea i don't think all the issues are completely solved, it's still moving very quickly, but i don't think i can discuss too much more
22:04 purpleidea but i do think there's a lot of cool tech there. does it fit *every* use case and need at the moment? of course not
22:04 JoeJulian Which is why I've stuck with openstack and kvm. Fits every use case for now.
22:06 PatNarciso I never got into openstack-- only because I don't know each component under it; and thuoght I'd spend more time figuring out how to fix something...
22:06 glusterbot PatNarciso: openstack's karma is now -1
22:06 PatNarciso sorry gusterbot... I'll learn not to do that.
22:07 PatNarciso so-- I've continue to rock kvm's manually.
22:07 glusterbot PatNarciso: so's karma is now -1
22:07 PatNarciso ...
22:07 PatNarciso is this something I need to get over, and dive into openstack OR is my old manual kvm editing ways OK?
22:08 purpleidea lol
22:08 PatNarciso glusterbot: openstack++ karma++ favor++
22:08 glusterbot PatNarciso: openstack++ karma++ favor's karma is now 1
22:08 PatNarciso argh.
22:09 purpleidea PatNarciso: you learn quick
22:09 JoeJulian I like using openstack as an abstraction layer for ease of administration.
22:09 LebedevRI joined #gluster
22:09 JoeJulian karma--
22:09 glusterbot JoeJulian: karma's karma is now -1
22:09 purpleidea PatNarciso: I think the people i work for offer openstack services, but i'm not an expert in that domain, nor a marketing/sales dude. but i think they're a good bet to talk to if you're interested in openstack
22:10 purpleidea we've got a lot of smart hackers *cough* *cough* </shameless plug>
22:10 PatNarciso have you ever found yourself wanting to do something outside of the openstack easy admin offerings?
22:10 purpleidea JoeJulian: i think he means you, not me
22:11 PatNarciso ex... I dunno, wanted to add a new openvpn bridge to a remote cluster?
22:12 JoeJulian That's in there.
22:13 JoeJulian The biggest problem with openstack it that it's huge and can do pretty much anything. So there's no good documentation that says, "do this" because if there was there would be thousands of "do this".
22:14 JoeJulian On the plus side, it's python. So if it's not working the way you want, change it.
22:14 dgandhi joined #gluster
22:14 PatNarciso gotcha.  looks like I need to carve a weekend out to playing with openstack.
22:15 JoeJulian .. s/a weekend/several weekends/
22:15 JoeJulian If you can learn it all in a weekend, I'll be extremely impressed.
22:20 PatNarciso understood.  I need more hours in a day.
22:34 badone joined #gluster
22:50 MacWinner joined #gluster
23:18 JoeJulian Gluster really doesn't like having a brick 100% full with a sparsely created file that's expanded to the full brick on one side of a replica but not the other.

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