[00:13] *** phreak has quit IRC (Quit: sleep) [00:24] *** EhtyarWRK has quit IRC (Quit: Never look down on someone unless you're helping them up.) [01:58] *** ziyourenxiang has joined #arpnetworks [03:02] *** milki has quit IRC (Ping timeout: 246 seconds) [03:19] *** ziyourenxiang has quit IRC (Quit: ziyourenxiang) [04:17] *** milki has joined #arpnetworks [04:39] *** Ehtyar has joined #arpnetworks [04:43] *** Ehtyar1 has joined #arpnetworks [05:02] *** Ehtyar has quit IRC (Remote host closed the connection) [05:28] *** milki has quit IRC (Ping timeout: 240 seconds) [05:34] *** milki has joined #arpnetworks [07:47] *** phreak has joined #arpnetworks [08:28] weird - getting some net breakage not to here, but to other sites [08:28] traceroute from laptop to delta.com stops at gblx.net for example [08:29] *** ziyourenxiang has joined #arpnetworks [09:11] *** ivan-kanis has joined #arpnetworks [09:17] *** ziyourenxiang has quit IRC (Quit: ziyourenxiang) [09:20] *** phreak has quit IRC (Quit: sleep) [09:34] *** HighJinx has quit IRC (Quit: Leaving) [09:53] *** phreak has joined #arpnetworks [09:54] *** cubelogic has joined #arpnetworks [10:05] --glbx. [10:05] been giving me hell all week at a couple of sites [10:05] (i.e., my connectivity output through them has sucked) [10:06] *** RandalSchwartz has quit IRC (*.net *.split) [10:06] *** bitslip has quit IRC (*.net *.split) [10:06] *** nukefree has quit IRC (*.net *.split) [10:06] *** twobithacker has quit IRC (*.net *.split) [10:06] *** Yamazaki-kun has quit IRC (*.net *.split) [10:06] *** hycer has quit IRC (*.net *.split) [10:06] *** up_the_irons has quit IRC (*.net *.split) [10:06] *** mjp has quit IRC (*.net *.split) [10:06] *** toddf has quit IRC (*.net *.split) [10:06] *** ix33 has quit IRC (*.net *.split) [10:06] *** azmarco has quit IRC (*.net *.split) [10:06] *** infrared has quit IRC (*.net *.split) [10:07] *** RandalSchwartz has joined #arpnetworks [10:07] *** bitslip has joined #arpnetworks [10:07] *** nukefree has joined #arpnetworks [10:07] *** twobithacker has joined #arpnetworks [10:07] *** Yamazaki-kun has joined #arpnetworks [10:07] *** hycer has joined #arpnetworks [10:07] *** up_the_irons has joined #arpnetworks [10:07] *** mjp has joined #arpnetworks [10:07] *** toddf has joined #arpnetworks [10:07] *** ix33 has joined #arpnetworks [10:07] *** azmarco has joined #arpnetworks [10:07] *** infrared has joined #arpnetworks [10:07] *** kornbluth.freenode.net sets mode: +oo up_the_irons toddf [10:10] *** phreak has quit IRC (Quit: sleep) [10:14] *** ivan-kanis has quit IRC (Remote host closed the connection) [10:22] *** HighJinx has joined #arpnetworks [10:39] *** fink has joined #arpnetworks [10:48] *** ww has joined #arpnetworks [10:51] *** ww has quit IRC (Client Quit) [11:09] *** fink has quit IRC (Ping timeout: 240 seconds) [11:11] *** fink has joined #arpnetworks [11:11] *** ww has joined #arpnetworks [11:12] up_the_irons: you work for arp, right and aren't just one of the clueful customers that hangs out in the channel? [11:12] it's true [11:12] he practically *is* arp. :) [11:13] up_the_irons: haaalp [11:13] RandalSchwartz: did you get an arp tshirt yet? [11:13] something we clueful people can help with? [11:13] fink - I don't think so [11:13] RandalSchwartz: me neither :( [11:13] haven't seen any recent packages [11:13] unfortunately not. [11:14] ww - then send email to support@ [11:14] it'll get done as fast as it can get done then. :) [11:14] machine has become unstable, probably because of zfs and insufficient ram (and i recently added some things that eat an extra 400Nb of ram) [11:14] so need a ram upgrade... erm urgently... [11:14] then you should probably back that out until you get more ram [11:15] i sent an email... [11:15] "urgent add of ram" is not a service provided here [11:15] "adding ram in a day or two" is more likely [11:16] you could temporarily add more swap space [11:16] *** ivan-kanis has joined #arpnetworks [11:17] that'd at least give you the option of pushing stuff out to disk that's less active. [11:17] backing out would be hard, in this case... spinning up another host elsewhere more likely if it'll take a while, but i'd rather not [11:18] poor planning then. :( [11:18] swap... well... 4s-backend (see 4store.org) gets very unhappy if it gets swapped (which happens now from time to time and is painful) [11:18] damn, 400Nb! [11:18] RandalSchwartz: planning? well... [11:18] Nanobytes! [11:18] nebibyte? [11:18] fink: 400Mb out of 1024 is... well... significant [11:18] ww: out of curiosity, what did you add? [11:19] i switched from apache to cherokee, and saved a shload of ram [11:19] actually, added nothing substantial new in terms of software (some minor upgrades) [11:19] what i did add was a pile of data [11:19] ... Optimum price/performance can be got with clusters of machines with large amounts of RAM. 16+GB per node is typical currently. [11:19] you mean you didn't read that and go "heck, I need more ram"> [11:19] that'd be my first clue [11:20] RandalSchwartz: depends on the quantity of data, actually [11:20] yeah - but anytime people say "this will need a lot of ram" what they mean is "I don't code very well" :) [11:20] with 16Gb of Ram you're talking ~100-150 million triples (rows, approximately) [11:20] i've only about 10 million [11:20] so why can't you just get rid of your data for now [11:20] roll it back, and wait for ram? [11:21] you didn't go live without testing did you? [11:21] because of a dog and pony show tomorrow [11:21] here we go… [11:21] again - poor planning. :( [11:21] what me plan? [11:21] ;) [11:21] remind me not to subcontract to you. :) [11:21] i DO have that tshirt [11:21] :D [11:22] oh, not a paying customer dog and pony show [11:22] if it were a paying customer they would be paying for bigger hardware [11:22] you missed the point, but that's fine [11:22] everyone else here got it. :) [11:26] anyways, re: planning... plenty of time to spin up a machine elsewhere if this can't be sorted quickly [11:27] I don't know why you seem to think that answers or contradicts my claim [11:27] ww: what OS? [11:28] because... there is no problem that can be attributed to inadequate planning? [11:28] fink freebsd 8.1 [11:29] and i'm *assuming* that Fatal trap 9: general protection fault while in kernel mode [11:29] is caused by ZFS WARNING: Recommended minimum kmem_size is 512MB; expect unstable behavior. [11:30] ww: i run fbsd 8.1 with zfs on root on a 768 vps, but i tweaked the zfs settings a bit, and i doubt i have your load [11:30] what processes / data are you running? [11:31] misc small things - it's the laboratory host [11:31] and, the metadata aggregator [11:32] which i would point you to, if the machine wasn't thrashing around again at the moment [11:34] ww: ok, what's your largest process group, maybe we can think of ways to make it leaner? [11:35] 4s-backend [11:35] not much to be done really. it alone occupies half the ram. [11:36] and there's an obligatory 4s-httpd whose size depends on the complexity of queries [11:36] then there's nginx, and a harvesting daemon [11:37] most everything else has been turned off [11:38] * ww just spins up a new machine... [11:42] *** fink has quit IRC (Quit: fink) [11:46] *** fink has joined #arpnetworks [11:57] *** phreak has joined #arpnetworks [12:17] fink: mind sharing some of the tweaking you did to run ZFS on a 768VPS? I considered it but figured that it'd freak out due to lack of ARC space if I used ~650mb of that 768 for actual running processes [12:19] * ww is interested in that as well [12:39] ... http://wiki.freebsd.org/ZFSTuningGuide [13:36] for interest's sake, the service in question http://semantic.ckan.net/ (pretty pictures here, http://semantic.ckan.net/record/2ab28263-a1fe-4839-9409-fa7ef03e123e and here http://semantic.ckan.net/group/?group=http://ckan.net/group/lld) is noww back up and running smoothly [13:36] unfortunately, elsewhere (rackspace) [13:55] *** ivan-kanis has quit IRC (Remote host closed the connection) [13:57] elastichosts is a lot more flexible than rackspace [13:58] at least in my observation [13:58] we're using elastichosts for BCP behind ARP [13:58] with elastichosts, if you want more ram, you slide a slider, and reboot. :) [13:58] it's *that* fast [14:05] I'm building a 8.2 box on elastichosts right now [14:06] yes, i'm familiar with them, but there's already an account, etc. set up with rs... [14:19] randalschwartz: elastichosts.com/could-hosting/pricing clearly shows that their smallest server is 1gb mem .. if you do a 'subscription' for 1gb mem 40gb disk 400gb xfer (same as arp's $30/mo system) you'd pay $107.92/mo there. huge difference in my world. [14:20] though as BCP plans go, it surely is good to make them [14:22] that doesn't make sense [14:22] I also don't note IPv6 mentioned, perhaps you could answer if there is remote encrypted console access and/or serial console access [14:22] I was dialing in some plans at $70/mo [14:22] so the minimum can't be 107.92 [14:22] randalschwartz: minimum isn't 107.92 [14:23] its just that if you compare apples to apples their minimum mem is 1gb [14:23] finding the 1gb mem option at arpnetworks, I chose the values from arpnetworks and plugged it into their little slider pricing guide [14:23] ... https://sat-p.elastichosts.com/docs/pricing [14:24] yeah - it's a bit more expensive overall [14:24] nothing is as cheap as ARP :) [14:24] and yeah, no ipv6 yet [14:24] they say "just use a tunnel" :( [14:24] "let he.net deal with IPv6 till we get our heads out of the sand" [14:25] indeed. anyone not dealing with ipv6 today has their heads firmly hidden :) [14:26] I've seen some 4U plans for 100mbit links unmetered if I provided the server. so after it approaches $100 for finite resources i tend to gloss over and out [14:27] "plans" => for $100 [14:37] toddf: I'm ready for OpenBSD 4.9 BTW, so whenever you want to make those templates... [14:41] ack [14:42] ahh - so you *can* use freebsd-update to go from 8.1 to 8.2 without recopying the boot blocks [14:42] unlike 8.0 to 8.1 [14:42] for zfs-on-root [14:42] just verified this in a private vm [14:57] nice [14:59] so that means I can upgrade my 4 VPS with minimal steps [14:59] everything at the # prompt :) [15:56] * dferris <3 the Dell iDrac 6 [16:02] up_the_irons: tell me how to make templates. I'll make you one for CentOS 5.6 (and 6.0 if they ever release it) and RHEL 5.6/6.0 [16:06] *** EhtyarWRK has joined #arpnetworks [16:14] jpalmer: can you do FreeBSD 8.2 as well? I can write-up the docs tonight [16:16] up_the_irons: I'd love to. [16:16] I can likely do FreeBSD this week. are you interested in centos/rhel? [16:17] yeay freebsd :) [16:18] and, can I get away from the single / partition? like, the sysinstall defaults? :) [16:22] jpalmer: I think that single partition on these relatively small VPSs is probably preferable? [16:22] single partition is NEVER preferable. [16:23] (but, I'll do whatever garry wants) [16:23] jpalmer: ummm strictly speaking, Red Hat likely wouldn't like the idea of a RHEL VM image if there is now VM/Cloud licensing [16:23] *no [16:24] G good point. they'd want licensing. so CentOS only [16:24] that said.... [16:24] jpalmer: due to the different access patterns on the different POSIX dirs, or something else I'm not thinking of? [16:24] I guess you could label it as bring your own license, no updates otherwise [16:24] reardencode: filling up you / is a lot more dangerous than filling up most other mountpoints. [16:27] reardencode: but that aside, a small / is preferable, because all of the other paritions can be background fsck'd. / has to be done at boot. in the case of an unclean mount, a 20g / takes longer to fsck than a 512m / [16:28] jpalmer: both valid points -- I guess you'd probably have to do at least two different templates, one for 5G vps and another for any other size, in order to leave even a reasonable amount of space for /usr on the 5G [16:30] a single ZFS partition is fine [16:31] because I can rehack it at will [16:31] personally, I'm ok with whatever garry wants for a template. I can always resize on my own. [16:31] really as ZFS matures, the many filesystem roots in a single pool of storage will win [16:31] RandalSchwartz: yep. [16:31] so if you could also offer /-as-ZFS as a template, that'd be great [16:31] I doubt most people will want to run ZFS on a vps due to the limited RAM resources. [16:32] and yet, I have four of them [16:32] and it's all working fine [16:32] also, to this day, I've never messed with ZFS (I don't have a need for it) so, I don't think I'd be the right guy to build a ZFS-as-root template [16:32] RandalSchwartz: guess we're not most people [16:32] apparently [16:32] we don't run in jpalmer's circles :) [16:33] heh [16:33] my circles are small, I chase my own tail a lot. [16:33] I chase tail a lot too. [16:33] oh - "my own" no [16:45] RandalSchwartz: how much RAM do you use for programs of your 768MiB with ZFS? [16:50] I have to think which one of the boxes are like that. [16:50] probably mabel. lemme look [16:51] says 278MB free [16:51] that box is just email and DNS [16:51] What do you have arc and kmem tuned to? [16:52] the recommedations on ZFSTuningGuide [16:52] arc = 40M [16:52] kmem = 330M [16:53] search for 768 in http://wiki.freebsd.org/ZFSTuningGuide [16:53] I'm using those [16:55] haven't had a problem in 10 months or so [16:55] even with a fair amount of spam [16:56] running postfix with postgrey and amavisd [17:00] *** jazz57 has joined #arpnetworks [17:17] amavisd is such a memory hog! [17:17] i have two vpses with the same setup as RandalSchwartzDoppleGanger [17:28] *** jazz57 has left [17:32] up_the_irons here? [17:35] jaja - why? [17:36] I dont see the payment on my CC, I want to know if there's a problm. [17:38] jaja: it can take a couple days [17:38] up_the_irons: second VPS requires the /29 subnet right? [17:38] jaja: wait…payment or charge? [17:39] is this a soviet russia joke? [17:39] arp pays you? [17:39] jaja: I don't think the billing cycle has happened yet [17:39] charge sorry [17:39] I am french my english isn't perfect, and I am abit tired tonight [17:39] fink: payment is fine imo, because he's making a payment to arp [17:40] you won't get your service disabled until he calls you [17:40] nothing to worry about [17:40] so if there's a problem with payment, it won't matter until he gets around to turning it off :) [17:40] just wait for the thugs in the black suits to appear :) [17:40] and the helicopters! [17:40] don't forget the helicopters! [17:41] RandalSchwartz: I thought he replaced the helicopters w/ drones [17:41] with the silent blades [17:41] I prefer fellow my billing, I am here since 6 month and generally charge is done 1th day of each month, sorry ! [17:41] whooop whooop whooop [17:41] all of the sudden they dart out of nowhere and pin you to the nearest brick wall :) [17:59] G: yes [18:26] jpalmer: OK, I'll write up the docs tonight. I don't have any templates for centos/rhel, so i can't give you a description of what is needed; you'd have to "figure it out" based on the Debian / Ubuntu template creation docs I have, and b/c the network is different, you might not want to go down that road [18:31] *** cubelogic has quit IRC (Ping timeout: 248 seconds) [18:46] up_the_irons: jpalmer: if help is needed regarding CentOS let me know [18:46] that said, I'm running Ubuntu on my VPS atm :) [18:55] *** HighJinx has quit IRC (Ping timeout: 240 seconds) [18:59] up_the_irons: nice. I'm setting up a pretty large bacula environment over the next 2-3 days. but, after that.. fbsd 8.2 template! [19:00] jpalmer: sweet! [19:06] *** EhtyarWRK has quit IRC (Remote host closed the connection) [19:06] *** EhtyarWRK has joined #arpnetworks [19:08] up_the_irons: support ticket best for second VPS I take it right? [19:10] *** EhtyarWRK has quit IRC (Client Quit) [19:11] *** EhtyarWRK has joined #arpnetworks [19:11] *** EhtyarWRK has quit IRC (Remote host closed the connection) [19:14] *** EhtyarWRK has joined #arpnetworks [19:14] up_the_irons: actually, if I only needed IPv6 on my second VPS, would you be able to do that w/o a /29? [19:25] G: yes, i would in fact [19:25] G: and yeah, support ticket for 2nd vps [19:25] up_the_irons: okay, sent [19:25] roger [19:26] up_the_irons: I only need IPv4 very occassionally, so if I can just use a private IP, and route the occasional packet via VPN or over the private network, then it save the pool and everything [19:26] I noticed the other day that American IP's a worth more than Asian IPs [19:27] *are [19:27] lol [19:28] up_the_irons: APNIC have a crazy formula for working out the costs of IP address assignment,s but I did the math and yeah, for big assignments APNIC look cheaper, kinda crazy in a way I guess [19:31] i c [19:31] so IP addresses do in fact have a real dollar value now? [19:31] interesting [19:31] of course, they're not marketable [19:31] since you can't reassign without the NICs getting involed [19:32] RandalSchwartz: well, there is the membership fees etc [19:33] RandalSchwartz: ARIN look to charge by the size of the assignment on a broadbasis, APNIC charge essentially on a per-ip basis [19:33] I actually had to remember how to do log's with the APNIC pricing [20:16] *** fink has quit IRC (Quit: fink) [20:19] *** EhtyarWRK has quit IRC (Remote host closed the connection) [20:23] *** HighJinx has joined #arpnetworks [20:27] *** EhtyarWRK has joined #arpnetworks [21:05] *** KyrosKoh has quit IRC (Read error: Connection reset by peer) [21:05] *** KyrosKoh has joined #arpnetworks [21:42] *** CRowen has quit IRC (Read error: Connection reset by peer) [21:47] *** CRowen has joined #arpnetworks [22:22] *** ivan-kanis has joined #arpnetworks [23:41] *** ivan-kanis has quit IRC (Remote host closed the connection)