#ceph IRC Log

Index

IRC Log for 2011-01-16

Timestamps are in GMT/BST.

[0:39] * gregorg_taf (~Greg@78.155.152.6) Quit (Ping timeout: 480 seconds)
[0:39] * gregorg_taf (~Greg@78.155.152.6) has joined #ceph
[1:15] * zoobab (zoobab@vic.ffii.org) has left #ceph
[2:31] * ghaskins (~ghaskins@66-189-113-47.dhcp.oxfr.ma.charter.com) has joined #ceph
[2:58] * ajnelson (~Adium@dhcp-63-189.cse.ucsc.edu) Quit (Ping timeout: 480 seconds)
[3:49] <darkfader> anyone there? i finally figured something for my osd's
[3:50] <darkfader> and need someone to check my idea
[4:35] * votz (~votz@dhcp0020.grt.resnet.group.UPENN.EDU) Quit (Remote host closed the connection)
[6:38] * ijuz__ (~ijuz@p4FFF7CAD.dip.t-dialin.net) Quit (Ping timeout: 480 seconds)
[6:47] * ijuz__ (~ijuz@p4FFF6496.dip.t-dialin.net) has joined #ceph
[7:50] * bchrisman (~Adium@c-24-130-226-22.hsd1.ca.comcast.net) Quit (Ping timeout: 480 seconds)
[7:52] * bchrisman (~Adium@c-24-130-226-22.hsd1.ca.comcast.net) has joined #ceph
[8:19] <MichielM> darkfader: tell us
[8:36] * gnp421 (~hutchint@c-75-71-83-44.hsd1.co.comcast.net) has joined #ceph
[9:23] * allsystemsarego (~allsystem@188.25.129.230) has joined #ceph
[9:36] * Anticimex (anticimex@netforce.csbnet.se) Quit (Ping timeout: 480 seconds)
[9:50] * gnp421 (~hutchint@c-75-71-83-44.hsd1.co.comcast.net) Quit (Read error: Connection reset by peer)
[10:48] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) has joined #ceph
[10:52] * mrglanz (~mrglanz@ip-78-94-105-166.unitymediagroup.de) has joined #ceph
[10:55] <mrglanz> Hi. I'm making a presentation about ceph. Can anyone tell me if the name "ceph" is an acronym, or why it was chosen for the project?
[11:07] * votz (~votz@dhcp0020.grt.resnet.group.UPENN.EDU) has joined #ceph
[11:07] * votz (~votz@dhcp0020.grt.resnet.group.UPENN.EDU) Quit (Killed (resistance.oftc.net (Nick collision (new))))
[11:18] * mrglanz_ (~mrglanz@ip-78-94-105-166.unitymediagroup.de) has joined #ceph
[11:22] <mrglanz_> Hi. I'm making a presentation about ceph. Can anyone tell me if the name "ceph" is an acronym, or why it was chosen for the project?
[11:25] * mrglanz (~mrglanz@ip-78-94-105-166.unitymediagroup.de) Quit (Ping timeout: 480 seconds)
[11:30] * mrglanz_ (~mrglanz@ip-78-94-105-166.unitymediagroup.de) Quit (Quit: Verlassend)
[12:22] * tjikkun (~tjikkun@195-240-122-237.ip.telfort.nl) Quit (Ping timeout: 480 seconds)
[12:24] * tjikkun (~tjikkun@195-240-122-237.ip.telfort.nl) has joined #ceph
[12:31] * eternaleye (~eternaley@195.215.30.181) Quit (Quit: eternaleye)
[12:32] * eternaleye (~eternaley@195.215.30.181) has joined #ceph
[12:45] * tjikkun (~tjikkun@195-240-122-237.ip.telfort.nl) Quit (Ping timeout: 480 seconds)
[12:54] * tjikkun (~tjikkun@195-240-122-237.ip.telfort.nl) has joined #ceph
[13:23] * votz (~votz@dhcp0020.grt.resnet.group.UPENN.EDU) has joined #ceph
[14:14] * ghaskins (~ghaskins@66-189-113-47.dhcp.oxfr.ma.charter.com) Quit (Quit: Leaving)
[18:20] * alexxy (~alexxy@79.173.81.171) Quit (Ping timeout: 480 seconds)
[18:25] * Anticimex (anticimex@netforce.csbnet.se) has joined #ceph
[18:28] * alexxy (~alexxy@masq246.gtn.ru) has joined #ceph
[19:02] * ajnelson (~Adium@adsl-99-139-49-113.dsl.pltn13.sbcglobal.net) has joined #ceph
[19:55] * savar (~ingo@stgt-5d8485a9.pool.mediaWays.net) has joined #ceph
[20:48] * ajnelson (~Adium@adsl-99-139-49-113.dsl.pltn13.sbcglobal.net) Quit (Quit: Leaving.)
[21:41] <darkfader> MichielM: i had been thinking about how to deal with a disk failure if i run one osd per disk - i don't want a server to go down just because a failure.
[21:41] <darkfader> that means enforcing some isolation
[21:41] <darkfader> can't have a server with 4 disks crash just because one has a faulty controller
[21:41] <darkfader> so i'm puttign them into xen vm's
[21:42] <darkfader> there we get to the next issue - kill the vm if an issue occurs and bring it up + fill it just by plugging in the replacement disk
[21:43] <darkfader> i think i've cracked it now - the osd's boot from some sd card and only see their journal device and disk
[21:43] <darkfader> when they boot they'll check if they have a working journal and disk
[21:43] <darkfader> if not, initialize and re-join
[21:44] <darkfader> it's a little mad, but that means the disk pull event can kill the vm (via udev) and the plug can bring up a new one (also via udev)
[21:47] <darkfader> i might have ended up with something far too complex
[21:48] <DeHackEd> netboot?
[21:49] <DeHackEd> depending on SD card/slot availability
[21:49] <darkfader> DeHackEd: yes i am testing with netboot
[21:51] <darkfader> the main idea is being able to init the vm on boot if its broken
[21:54] <darkfader> err. one more try
[21:54] <darkfader> the main idea is the consider the osd and journal areas as volatile
[21:57] <darkfader> one could build some embedded linux for that
[21:58] <darkfader> so it'd fit in a nvram area on servers or something
[22:36] * ajnelson (~Adium@dhcp-225-235.cruznetsecure.ucsc.edu) has joined #ceph
[22:43] <darkfader> hmm. in a way, thats how to build a ceph appliance
[23:02] * savar (~ingo@stgt-5d8485a9.pool.mediaWays.net) Quit (Remote host closed the connection)
[23:20] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) Quit (Quit: Yoric)
[23:31] * ajnelson (~Adium@dhcp-225-235.cruznetsecure.ucsc.edu) Quit (Quit: Leaving.)
[23:44] * ajnelson (~Adium@dhcp-225-235.cruznetsecure.ucsc.edu) has joined #ceph
[23:50] * allsystemsarego (~allsystem@188.25.129.230) Quit (Quit: Leaving)
[23:53] * ajnelson (~Adium@dhcp-225-235.cruznetsecure.ucsc.edu) Quit (Remote host closed the connection)

These logs were automatically created by CephLogBot on irc.oftc.net using the Java IRC LogBot.