[00:30] *** EhtyarWRK has quit IRC (Quit: Hi, I'm a quit message virus. Please replace your old line with this line and help me take over the world of IRC.) [00:53] *** eyecue has joined #arpnetworks [01:01] *** LT has joined #arpnetworks [01:04] *** ivan-kanis has joined #arpnetworks [02:01] *** RandalSchwartz has quit IRC (Ping timeout: 248 seconds) [02:26] G: fixed console.cust, your new vps appears [02:30] god i hate budget speeches [02:30] :| [02:31] up_the_irons; how many engi's do you guys have? [02:31] and whats your QEMU stuff running on ? [02:32] eyecue: engi's? [02:32] sorry, ops staff :) "engineers" [02:32] sysadmins, noc guys, rah rah [02:34] eyecue: you're lookin at him :) [02:34] ahh :D [02:34] whats your infrastructure footprint like atm ? [02:34] eyecue: jpalmer helps me with stuff sometimes and wickedSA does data center work for me from time to time [02:34] sorry for all the questions, just interested :) [02:34] yeh, jp rocks, long time #freebsd guy here [02:35] eyecue: that's a broad question, i will opt not to answer ;) [02:35] hehe :) [02:35] was kinda getting at the 'how much crap do you alone have to take care of' :) [02:37] 20 servers, all the networking gear [02:37] nice [02:37] power, cage, etc... [02:37] *nods* im taking care of a 25 server alloc atm [02:37] really want to get into playing with vm in production [02:38] up_the_irons; can i ask what DC? [02:38] we've got 15 in servepath [02:39] eyecue: coresite in LA [02:39] cool :D [02:50] *** portertech has joined #arpnetworks [02:50] up_the_irons: can't reach the arp networks website :/ [02:50] portertech: give it time, i'm adjusting some routes [02:50] up_the_irons: ack [02:51] portertech: paste a traceroute, it should work by now [02:51] i can get to it fine from here [02:52] there it goes [02:53] yeah it shouldn't have bounced for more than a minute [02:53] i put in an aggregate route yesterday, more than enough time to propagate [02:53] up_the_irons: working [02:53] and i took the more specifics out just now [02:53] (japan) [02:53] portertech: cool [02:53] redundant network and upstream links ? [02:54] eyecue: you ask as if there is an alternative? ;) [02:54] alternative/ideal/bestpractice and reality are often two different things :d [02:56] up_the_irons: is the dashboard a sinatra app? [02:57] portertech: no, the main website is sinatra, the portal is Rails [02:57] eyecue: indeed, true [02:57] up_the_irons: neat [02:57] yeah [02:57] * eyecue mumbles something about django [02:57] :D [03:01] *** milki has quit IRC (Ping timeout: 240 seconds) [03:03] up_the_irons: yay, I'm an arp networks customer once again! :P [03:03] just sent my order [03:03] portertech: nice; how is the latency from japan? [03:04] currently surf'n on 4G, not really a good indication, I'll know more once I have an arp vps provisioned [03:05] ah ok [03:05] up_the_irons; is there an alternative to 95th percentile billing and/or volume bandwidth across multiple servers? [03:05] eyecue: as in? those two are all i've ever heard of [03:07] up_the_irons; two questions, can you do average usage billing, and second, can i purchase a block of bandwidth as a seperate line item to utilise across multiple servers, instead of a XXXgb block per server [03:09] up_the_irons; rationale behind the question, we currently have 15 servers, 5 of which run high network utilisation workloads for 6 hours a day, and 0 for the rest of it, and the other 10 run more evently distributed network loads [03:09] eyecue: bandwidth is already a separate line item (1 vps can use all of it, and the others can use none, etc...). i don't do average usage (other than 95th) [03:09] eyecue: which 6 hours of the day? :) [03:10] 95th percentile works out to almost twice as expensive for us for the servers that burst for that short period a day, and the bandwidth that comes along with the other 10, goes under utilised over the month [03:10] up_the_irons; umm, nighttime at gmt+10 [03:10] how high is the utilization? [03:10] *looks at zabbix [03:11] brb im going to drop :D [03:11] vpn fail [03:13] 15-20Mbps per server x 4 in/out [03:13] sometimes 30-50% up on that [03:14] up_the_irons: thakns, I'll check it out soon [03:16] G: np [03:16] eyecue: i c [03:16] :] [03:17] eyecue: i'm low on space for colo, but i think someone that would be able to accommodate that is Syminet: http://syminet.com. The owner is Mark Symonds and he's a real cool guy. He buys bandwidth from me too, so it is same quality that I offer. [03:17] up_the_irons; even vps's ? [03:18] or straight physical colo only ? [03:18] eyecue: oh, we're talking vps' here? thought you meant physical servers [03:18] hopefully recomming some physicals as vps's [03:18] *** hycer has quit IRC (Read error: Operation timed out) [03:18] no point having a quad core around for 6 hours of usage [03:19] not very IO heavy workloads, so its primarily a cpu/ram thing [03:19] *** hycer has joined #arpnetworks [03:19] eyecue: i can handle the vps needs; but why do average billing for that, i could give you a bulk discount on total data transer (GB) usage [03:19] eyecue: yeah true [03:19] up_the_irons; sure, i was wondering about flexibility over the billing model [03:20] more so than the actual solution :) [03:20] i know how hard it is to change in contract :D [03:20] eyecue: generally, i'm flexible, if it makes sense. but i'd rather not create a whole new billing model if i don't have to ;) [03:20] absolutely, everyone needs to make moneys [03:21] in our .au deployment, we pay per month for the physical hardware, and buy a few tb of bandwidth to utilise coming into our distribution switches [03:21] up_the_irons; ahh heres a good one, do you charge 95th percentile on the highest of up/down utilisation, or over the sum ? [03:24] eyecue: cacti says: HRULE: |95:bits:0:max:2| [03:24] :] [03:24] i keep forgetting what that means ;) [03:24] hehe [03:24] i think it is the higher of the two [03:24] roger that [03:25] eyecue: where are you located? [03:26] sydney, au [03:26] deployments in bne, au, syd, au, and sfo, us [03:27] nice [03:28] ahhh, good ol' BNE :) [03:28] :] [03:28] I love that city, walking alongside the river etc [03:28] up_the_irons; currently got a large mail infrastructure in US, HA-WWW arch in sydney, and dns services in BNE [03:29] want to get back there sometime [03:29] G: not last year it wasnt (sure you saw the flood coverage) [03:29] one of the biggest DC's in the country got literally shutdown [03:29] eyecue: cool [03:29] eyecue: yeah, that flood... [03:29] wow [03:29] up_the_irons; exclusively freebsd [03:29] eyecue: sweet [03:29] eyecue: which DC was that? [03:29] G: AAPT iirc [03:29] ahhhh [03:30] are they still owned by Telecom NZ? or did Telecom finally offload it to TPG? [03:30] http://goo.gl/UNDIP [03:30] no idea [03:30] who knows who owns who these days [03:30] check out those google results though ;) [03:30] true I guess :) [03:31] eyecue: the bit of the CBD I used to work in was pretty much unaffected iirc [03:31] *nods* i had a server in the Pipe DC on creek st [03:31] it was up on a 45 degree street [03:31] Creek Street, which one is that again? [03:31] Pipe Networks [03:32] eyecue: no, I meant which street :) [03:32] I got the Pipe networks bit :) [03:32] which street? creek street :D [03:33] ohhhh next one over from Edward St [03:33] yeh, google maps was playing up there o_O [03:34] eyecue: Google Maps was giving me nowhere near the CBD [03:34] yeh same [03:34] Redcliff for instances [03:34] i even tried to redo the query [03:34] i got 4034 [03:34] where are you at at the moment ? [03:34] yep, I also got a bunch in the 4030's [03:34] eyecue: New Zealand [03:34] ah [03:34] eyecue: I moved back last year [03:35] hopefully im coming over to Mt Hutt soon for skiing [03:35] :D [03:35] never seen snow in my life, very excited [03:35] I used to work on the otherside of the CBD... North Quay/Herschel Street way [03:35] ive decided i dont like 95th percentile bandwidth models [03:35] :| [03:36] eyecue: do you get up to BNE much? [03:36] useto live there [03:36] moved back to syd a few years back [03:36] lived in stones corner (near the gabba) [03:36] have you seen the Kurilpa Bridge? [03:36] worked at park road milton [03:36] ahhh yep, know Park Road [03:37] hehe, tank street bridge :D [03:37] iirc thats where APNIC were based [03:37] yah [03:37] yup, they were downstairs from us @ APN :] [03:37] 33 park road *win* [03:37] eyecue: I generally refer to it as the Ugly Bridge :) [03:37] i asked one of the apnic guys in the elevator one day for a /8 :] [03:37] eyecue: iirc that's also the building QTAC is/were in right? [03:37] he laughed [03:37] yup [03:37] eyecue: haha, thats a great one [03:38] :) [03:38] Red Hat used to also be down Park Road [03:38] mm, news to me [03:38] how long ago ? [03:38] eyecue: ummm they shifted from there in 09 iirc [03:38] ah k [03:38] no wait, 08 [03:38] memories :] [03:38] I think it was Mid 08 [03:39] ithat was right about the time i was there, ish. [03:39] I interviewed at their new place in August/September-ish 08, and started working for them in Nov 08 [03:39] one of the guys i worked with @ APn took this photo of my dd and his best mate in front of la dolce vita :D [03:39] i can tell you exactly when we were there [03:40] 18 august 08 :] [03:40] waiting for linky [03:41] ahh here we go [03:41] http://keithrowell.com/365/2008-08-18.jpg [03:42] oh thats right, it wasn't directly on Park Rd, it was off a sidestreet [03:42] http://keithrowell.com/365/2008-09-17.jpg <-- this ones inside la dolce vita [03:43] eyecue: nice [03:43] :] [03:44] But yeah, that Tank Street bridge (imo) is Ugly :) [03:44] big time [03:44] eyecue: what made it worse, was I had to see it litterally every single day [03:44] :| [03:45] where were you working? [03:45] I'd say the only day I didn't see it, was the week I was couped up in bed sick as a dog :) [03:45] eyecue: North Quay/Herschel Street [03:45] kk [03:45] and I lived on Tank Street [03:45] wow [03:46] I had a very fast commute :) [03:46] 20 a month for a fbsd vps on arp, nice [03:46] except I used to walk the long way around the block to get decent coffee :P [03:46] :) [03:46] i drove in after 9:30 to avoid traffic [03:47] eyecue: I didn't have that luxury [03:47] eyecue: how did you hear about arp, btw? [03:47] up_the_irons; mentioned it earlier, @obfuscurity [03:47] up_the_irons: btw, liked your tweet about Slicehost, didn't realise they were migrating everyone to RS Cloud [03:47] uh, jason dixon [03:48] eyecue: oh ok [03:48] G: yeah i just heard about it a few days ago [03:48] up_the_irons; https://twitter.com/#!/obfuscurity/status/67602845507125248 [03:48] up_the_irons; i thought, if they do obsd, theyll do fbsd [03:49] eyecue: yep :) [03:49] freebsd is the most popular OS around here [03:49] yay:) [03:51] oo, 30loops using freebsd for their firewalls, cool. [03:52] portertech: so did you move to slicehost and then come back here? ;) [03:53] I have vps's all over the map (except rackcloud/slicehost atm :P) [03:53] nice [03:53] I've been meaning to drop a few and move back to arp [03:53] anyone know if you can use CARP on a single /30 subnet? I know which HSRP and VRRP, you have to jump through hoops [03:53] :) [03:54] why do you think you couldnt ? [03:54] 2 usable ? [03:54] up_the_irons: for what it's worth, before I asked you about the 2nd VPS, I signed up to Rack Cloud and took a look (really, I only need bursts of a second VPS every now and then, so I thought, all save heaps only pay a few dollars), I was mistaken :) [03:55] I could have done it, but it would have been ugly [03:55] G: hah [03:55] easier to just do it as a second VPS on ARP [03:55] nice :) [03:55] up_the_irons; each physical needs its own ip, plus the virtual ip [03:56] up_the_irons; thus at least a /29 [03:56] eyecue: it's the virtual IP that concerns me. so, you have device A and device B (2 IPs), a broadcast address (1 IP), and network number (1 IP). no room for virtual IP [03:57] correct [03:57] you could proxy arp's, so they could speak [03:57] but meh. [03:57] up_the_irons: they count powered down, time for costs, and the only way to keep the data w/o hourly charges is to move it to their files thing, and then delete the server (and get a new IP) etc [03:57] so then you can't do CARP on a /30 ;) [03:57] so meh [03:57] correct. ive read at some point that future versions may support it [03:57] G: that's dumb [03:57] but no specifics [03:57] i c [03:58] but carp does rock, majorly. [03:58] running lagg+carp for our HA inf in sydney [03:58] +haproxy+nginx (ssl) [03:58] HSRP on the upstreams and for the private lan [03:58] ahhh I see CARP is kinda HA... [03:59] G: its a patent free replacement for VRRP [03:59] fail cisco. [04:01] ah ha, yep, I recognise it now :) [04:02] now about that beta freebsd vps ;D [04:02] eyecue: Cisco seems to be the perfect example of "Do what I say, but not what I do" :) [04:02] mm, indeed. [04:03] the one thing I really like about Cisco though, is IOS [04:04] really? :) [04:04] eyecue: yeah, when they consistently apply the IOS style through devices, it's a godsend imo [04:05] fair call [04:05] i really wouldnt mind a small l3 cisco switch to play with [04:06] eyecue: I had to jump in the deep end w/ two of their Fibre Switches the other year, hadn't touched Cisco stuff for ages, but the basics stuck with me, and it all worked easy as pie [04:06] i did that with my first 2621XM with one of my old clients [04:06] (keeping in mind that I'd only dealt w/ Cisco switches and routers in the past as well) [04:06] Ethernet that is [04:06] howd you find it ? [04:07] eyecue: easy as pie [04:08] eyecue: sure I had to look at the documentation to understand the specifics, but as they keep the basic syntax and concepts/style the same, it was pretty easy [04:08] of course, but looking at docs isnt a bad thing [04:09] in the same vein as being able to regurg doco isnt necessarily a good thing :) [04:18] eyecue: sounds about right [04:21] *** milki has joined #arpnetworks [04:59] *** ivan-kanis has quit IRC (Remote host closed the connection) [05:04] *** CRowen has quit IRC (Read error: Connection reset by peer) [05:13] *** CRowen has joined #arpnetworks [06:27] up_the_irons: off the top of my head, i think you'll need a 4000/4500/5500/6500 to get that many vlans [06:30] unless you go non-cisco, of course [06:31] oh dear, MS gets skype. [06:32] eyecue: as I just said in another channel, hope Facebook is their next target [06:33] :| [06:33] that aint gonna happen :) [06:33] sooner trust Microsoft w/ my personal data than Zuckerberg [06:33] eyecue: they already own a small portion :) [06:33] dont have a FB account :) [06:33] so i dont have that problem [06:33] yeah, I've been meaning to delete mine for ages [06:36] time for another glass of red [06:36] bbs [07:59] *** portertech_ has joined #arpnetworks [08:01] *** portertech has quit IRC (Ping timeout: 240 seconds) [08:01] *** portertech_ is now known as portertech [08:10] thinking maybe i should use nginx as a frontend instead of as complete apache replacement, always such a pain to set sites up with things all ready to roll on apache [09:05] *** mattx86 has joined #arpnetworks [09:16] *** LT has quit IRC (Quit: Leaving) [09:45] *** cubelogic has joined #arpnetworks [09:46] *** fink has joined #arpnetworks [09:48] *** HighJinx has quit IRC (Ping timeout: 240 seconds) [10:21] *** HighJinx has joined #arpnetworks [11:58] *** Sheath has quit IRC (Quit: ZNC - http://znc.sourceforge.net) [12:04] *** Sheath has joined #arpnetworks [12:04] *** Sheath has quit IRC (Changing host) [12:04] *** Sheath has joined #arpnetworks [12:20] *** danzasph1re has joined #arpnetworks [12:20] *** danzasph1re has quit IRC (Client Quit) [12:21] *** PatSphere has joined #arpnetworks [12:24] *** heavysixer has quit IRC (Quit: BAMPF!) [12:33] *** Lefty_ has quit IRC (Disconnected by services) [12:36] *** PatSphere has quit IRC (Ping timeout: 264 seconds) [12:37] *** PatSphere has joined #arpnetworks [12:37] *** heavysixer has joined #arpnetworks [12:37] *** ChanServ sets mode: +o heavysixer [12:47] *** PatSphere has quit IRC (Ping timeout: 264 seconds) [12:48] *** PatSphere has joined #arpnetworks [13:00] *** PatSphere has quit IRC (Ping timeout: 260 seconds) [13:01] *** alainb has joined #arpnetworks [13:01] not that anyone here will be testing ariane's vmmap_sys diff at ARP but me, but I recommend you not do so, a system behaving as if it is 1mhz is not my cup of tea. [13:01] *** PatSphere has joined #arpnetworks [13:01] hello everyone [13:02] who do i have to talk to for getting extra ips ? [13:03] alainb: that would be up_the_irons if he's around, but generally those type of requests get shuffled to support@ anyway. prepare for $1/additional v4 IP. [13:07] ok well i just sent a email thanks anyway [13:12] *** PatSphere has quit IRC (Ping timeout: 252 seconds) [13:14] *** PatSphere has joined #arpnetworks [13:16] *** alainb has quit IRC (Quit: Lost terminal) [13:21] jlgaddis: yeah i was aware that 4500/6500 do it, but i'm looking for like a 1U solution that isn't $7K+ (4900) ;) [13:22] up_the_irons: you can use CARP on a single IP w/out an IP in the system itself, just expect the BACKUP system can't access the network .. but that should be qed [13:23] toddf: i'd put a management IP on the backup system on a separate interface, so accessing it wouldn't be a problem [13:24] or just do v6 for accessing, and v4 for legacy apps.. *grin* [13:25] or if your apps speak v6, don't worry about that /30 ;-) [13:25] (not quite clear what is being done, though, so some advice may be irrelevent) [13:26] *** PatSphere has quit IRC (Ping timeout: 276 seconds) [13:27] toddf: i want to set up some kind of HA (hrsp/vrrp/carp/etc) for customer gateways [13:28] *** PatSphere has joined #arpnetworks [13:29] up_the_irons: ah, then you want something like this: [13:30] * up_the_irons listens [13:31] * toddf is currently whittling at shellcode, seemed simpler in head than in reality, will be a moment [13:34] np [13:36] for h in /etc/hostname.vlan*; do id=${h##*vlan}; ( echo "vhid $id pass $(pwgen -s 64) carpdev vlan$id"; egrep "^description|^inet " $h ) > /etc/hostname.carp$id; grep -v "^inet " $h > $h.tmp && mv $h.tmp $h; done [13:37] in english, move your 'inet ' stuff from /etc/hostname.vlanX to /etc/hostname.carpX [13:37] specifying carpdev in /etc/hostname.carpX in reference to vlanX [13:38] sweet [13:39] note that if you want to do carp with IPv6 I'm currently trying to convince someone to do something about the fact that 'rtadvd carpX' will work, but 'ping6 -c 1 fe80::X%em0' is neccessary upon client's initial bootup or somehow packets don't get to the right place on the router. [13:39] wish i could use it, however, my primary side router is a cisco 4500, so only hsrp and vrrp are my options [13:40] *** PatSphere has quit IRC (Quit: Lost terminal) [13:41] (probably qed but 'rtadvd' takes a list of interfaces, don't run one instance per interface .. *grin*) [13:43] i c [13:43] which also makes me curious, tell me if this is asking too much info, but do you have anything non default on your openbsd v6 router? typical v6 routers respond to a ping6 of ff02::2%em0 but yours doesn't.. [13:47] *** PatSphere has joined #arpnetworks [13:50] I will try to reproduce with qemu -serial tcp:... but if anybody can point out what is required to send a break sequence over conserver to an ARP vps.. I'd be much obliged ;-) the ^Ecl0 seems not to be working for me.. [13:51] *** PatSphere has quit IRC (Read error: Connection reset by peer) [13:52] toddf: it's pretty default [13:52] I'm sure there's a reason. even if it has to do with loonix filtering something on the vlans (which I doubt, but you never know..) [13:53] toddf: nah, there should be no filtering at all on the vlans [13:54] pf.conf icmp filtering on the routers? [13:54] nope [13:55] its just .. odd. because any other system I've seen with net.inet6.ip6.forwarding=1 responds to ff02::2, but it clearly is forwarding packets... ;-) [13:55] random curious itch I'm scratching. no big deal, just .. odd. ;-) [13:56] yeah [13:59] *** portertech_ has joined #arpnetworks [14:01] *** portertech has quit IRC (Ping timeout: 240 seconds) [14:01] *** portertech_ is now known as portertech [14:02] odd, seems I have a local switch blocking something or something .. my laptop doesn't see traffic sent to ff02::2%fxp0 from a system on the same ethernet segment. [14:03] hehe [14:12] *** Pathin has joined #arpnetworks [14:13] *** Pathin has quit IRC (Read error: Connection reset by peer) [14:15] *** Pathin has joined #arpnetworks [14:18] *** milki has quit IRC (Ping timeout: 240 seconds) [14:20] *** Pathin has quit IRC (Read error: Connection reset by peer) [14:47] *** Pathin has joined #arpnetworks [14:47] up_the_irons: as much as i shit on hp, i quite like their stuff for layer 2 [14:49] jlgaddis: o'rly? [14:50] then maybe i'll check it out, cuz all i need is layer 2 [14:52] yeah, the layer 2 stuff is solid. just don't try to route anything. =) [14:53] *** Pathin has quit IRC (Read error: Connection reset by peer) [14:53] jlgaddis: ok :) [14:53] i'm not sure what list is on the e4210-48g, but i think that'd do what you need and should be ~$1800 or less list [14:53] awesome [14:54] i also got a suggestion for extreme networks x350 and an juniper EX4200-48P [14:54] so i got some homework to do [14:54] bet the juniper will burn my fucking bank down [14:55] easily 2x the hp [14:55] bankburner! [14:55] the ex4200 is nicer though [14:56] they're stackable, too, if that matters [14:56] virtual chassis shit [14:58] *** Pathin has joined #arpnetworks [14:58] jlgaddis: stackable may be nice when i exceed 48 vm hosts [14:58] i'm at 22 now, so it's not unlikely that'll happen within the next year or two [14:58] out of curiousity, how many physical... [14:58] ahh ok =) [14:59] :) [14:59] gonna need another cage [14:59] nah, i can go to at least 48 in the same space [14:59] i have an empty cab that can do 15 servers and then i might have a colo customer giving up a cab within the next year [15:00] they got sucked into "the cloud" [15:00] oh, one of those [15:00] srsly [15:10] *** Pathin has quit IRC (Ping timeout: 252 seconds) [15:12] *** Pathin has joined #arpnetworks [15:13] jlgaddis: what is the cli like on the hp's? is it "cisco-like" or... ? [15:18] "VLAN support and tagging: support IEEE [15:18] 802.1Q, with 4094 simultaneous VLAN IDs [15:19] " [15:19] yeah, so the HP can do what I want, cool [15:19] can't wait to replace this Dell PowerConnect 5324 [15:19] to be fair, it did its job for two years and was < $300 [15:20] *** Pathin has quit IRC (Quit: Lost terminal) [15:20] up_the_irons: not exactly like ios, but not majorly different. biggest thing you'll notice is how vlan's are configured. [15:21] jlgaddis: ok [15:21] i wonder if there is a rancid profile for HP [15:21] instead of 'conf t ; int ; switchport vlan access ' its 'vlan ; untagged ' [15:22] or 'vlan ; tagged ' instead of "switchport trunk ..." basically [15:22] yes, there is. i use rancid with all my procurves. [15:22] s/cisco/hp/ in the router.db file [15:23] jlgaddis: oh sweet, cool [15:24] *** phreak- has joined #arpnetworks [15:24] jlgaddis: (re rancid profile) [15:24] jlgaddis: that vlan config is like force10 [15:24] i don't have any force10, but a buddy of mine let me work on one once [15:24] yeah the old foundry's were the same way [15:25] i c [15:26] *** key has quit IRC (Read error: Operation timed out) [15:28] if you do end up getting an hp, ping me if you need anything. i, unfortunately, have way too much experience with them. [15:32] *** milki has joined #arpnetworks [15:39] jlgaddis: sure will do [15:46] jlgaddis: why are some blue-ish and some grey? i take it the grey ones are the older model? [15:46] grey/white that is [15:49] procurve? [15:49] the e4210's were 3com's originally [15:49] oh, which means the cli is different and they probably don't have a lifetime warranty [15:50] glad i just thought of that. not sure if that makes a difference to you or not. [15:50] jlgaddis: it actually does say lifetime warranty [15:51] why can't i find an f'in picture of one of these that is bigger than a thumbnail?! [15:54] up_the_irons: which model(s)? [15:55] pilgrimd: e4210-48g [15:58] up_the_irons: Is this large enough: http://h18000.www1.hp.com/products/quickspecs/PH_001737/PH_001737.JPG [15:59] pilgrimd: omg, yes, thank you! [15:59] that's not it [15:59] i have some of those in my office, i think they're 2510s [16:00] yeah looks like the 2510 [16:00] * pilgrimd can't see worth crap, it seems. [16:00] those guys are pretty weak it seems, only 64 vlans [16:01] thinking about it, i don't think any of the procurves (made by hp) can do 4094 vlan's. well, not 1u fixed-port layer 2 only ones. [16:01] just the stuff they got from 3com [16:02] *** eyecue has quit IRC (Ping timeout: 260 seconds) [16:02] those 4210s are probably the lowest end model that can [16:03] roger that [16:04] i wonder how good the ex-3com stuff is [16:04] haven't used any of it personally, but i've heard good things [16:04] if it were me and i could afford it, i would go with an ex4200 though [16:05] cisco > juniper > hp, imo =) [16:05] yeah :) [16:06] (vendor j)++ [16:07] *** Pathin has joined #arpnetworks [16:07] jesus the ex4200 is expensive (why am i surprised) [16:08] like $7K [16:08] talk to a sales guy and tell them it's between the ex4200 and a c4500 but you're "just not quite sure yet" and ask if he can do anything [16:08] juniper salespeople right now have a lot of leeway in throwing discounts out [16:09] (especially if you're a new customer) [16:09] oh i c [16:10] *** Pathin has quit IRC (Read error: Connection reset by peer) [16:10] found one at buy.com for $5K [16:10] $7k for a 48-port gig switch? sheesh. 3750X's are less than that. [16:11] *** Pathin has joined #arpnetworks [16:11] oh the ex4200 is L3 [16:11] don't need L3 [16:12]