[00:02] haha yeah [00:05] mhoran: the only reason 40 Gbps Ethernet became a thing was because the backplane in Cisco switches couldn't go higher than that. And aren't they a big player in IEEE ? So it was codified and everyone was like, " wtf fuck you".. because they held back adoption of 100 Gbps ethernet [00:06] mercutio: that Arista... it makes no sense to have 40x 100 Gbps but only 8x 40 Gbps.. the uplinks are a bottleneck [00:07] brycec: looks like I can ping one of your IPs now (yay!) [00:09] up_the_irons: that's why i thought it might be a typo it just seemed odd [00:09] but the datasheet matched [00:16] up_the_irons: What a wonder when systemd actually starts! [00:16] * brycec moves on to figuring out what the hell happened that Debian forgot how to /sbin/init [00:16] (much smoother now that I can SSH in) [00:17] that's always my worst fear after my arp box has been running for 2 years [00:17] 6 months in my case [00:17] Upside: zfs snapshots are kept for 6 months [00:17] I was shocked everything started back up without any intervention after the cage move the other day [00:18] You lucky duck :P [00:19] well well well... The March 1st snap is good, the April 1st snap has a busted /sbin/init [00:19] * brycec hugs zfs [00:19] wow how did /sbin/init get corrupted [00:19] hack? [00:19] mercutio: maybe the data sheet is wrong? hahha [00:19] Went from being a symlink to some old-ass sysvinit [00:20] (narrowed it down to sometime between March 1st and March 25th... because I only keep 30 daily snaps) [00:21] you think the result of an upgrade? [00:21] Must be? [00:21] (No evidence of unauthorized access) [00:22] that system must have been installed a long time ago if it has any sysvinit binaries anywhere near it :P [00:22] I mean, it's Debian, so... [00:25] heh [00:25] bit of a change of subject but [00:25] has ARP ever thought about buying ARM server hardware? [00:26] pretty niche atm, but aws is offering it now [00:26] Found a forum post (two, in fact) that described what happened precisely \o/ https://forum.proxmox.com/threads/upgrade-to-kernel-4-15-18-12-pve-goes-boom.52564/#post-243333 [00:26] might be cool to get in ahead [00:26] brycec: so it was zfs's fault in the end :P [00:27] (sort of) [00:27] shuddup :P [00:27] * brycec blames packages [00:27] *packagers [00:37] i wonder if the idea of the 8 uplinks is to connect to legacy 40gigabit networks for your 100 gigabit supercomputer cluster up_the_irons ? [00:38] "pve goes boom" hahahah [00:39] acf_: never thought much about it... i'm so tired of buying hardware [00:40] it's the opposite of an investment... if you want to make sure your money runs the other way of inflation, buy technology [00:41] I suppose it is a necessary evil though [00:41] buying hardware [00:41] https://www.gigabyte.com/ARM-Server/R150-T62-rev-10 [00:42] I wouldn't say buying is "necessary" - there's always theft :P [00:42] this looks like it's roughly US$5,000 [00:42] or, more legally, leasing :P [00:43] cool thing about these arm boxes is they seem to have a ton of cores [00:43] so I presume you can dedicate 1 core to each of 47 customer VMs [00:43] on a single blade [00:44] I've learned that to survive in this biz, you have to let someone else eat the depreciation. so, after AWS or Google offload 10,000 of these boxes onto eBay, then we'll buy some haha [00:45] hahaha fair enough [00:45] that's how I buy my laptops actually [00:45] always off ebay, and always at least 2 generations old [00:46] price is like 1/10th and the box is good as new [00:46] yup [00:46] I'm still using my Lenovo T520 which I got for $240 [00:47] I had another T520, which a client had bought me years earlier, and it was $1800 new [00:47] ah damn that's good [00:47] I was using an ebay X250 until recently [00:47] nice [00:47] I decided I needed 16GB for all my chromium tabs so I upgraded to a 5th gen X1 carbon, also off ebay [00:48] Not to mention by waiting, there's either an established market for $newtech or you know it's a flop and not to waste your money :) [00:48] spoiler: chromium figured out how to eat the 16gb [00:48] chrome is such a huge memory hog [00:48] i've started using firefox sometimes now [00:48] it seems to be getting worse and worse [00:49] computer cpus have hardly been rising in speed [00:49] they've actually been getting slower over time with all of these spectre/meltdown workarounds [00:49] https://unixcube.org/who/acf/tmp/images/screen/top-1.png [00:49] this pretty much describes it :P [00:50] there's very little performance boost from ddr3 to ddr4 [00:50] graphics cards have improved [00:50] chromium chromium chromium chromium chromium java chromium [00:50] nvme is stupidly fast for disk [00:50] but single threaded processing speed struggles to make much improvement [00:50] acf_: I have a tab addiction too [00:50] and most code still isn't designed to parallelise well [00:51] the main benefit to performance ordinary users get now is things like JIT compiling of javascript [00:51] I think it's time I leave the DC... it's been another like 12 hour day here [00:51] if people coded for performance like they did in amiga days computers would probably rocket along [00:51] yikes, yeah time to go home :P [00:52] :) [00:52] must be working quite hard with all this migration stuff recently [00:52] mercutio: haha yeah [00:52] acf: there's a lot of little things [00:52] acf_: certainly, it's been a whirlwind [00:52] but things are becoming simpler [00:52] You basically start over, but with a fresh design [00:52] ipv6 and ipv4 terminate on the same router [00:52] and you take what you learned over the years and make changes [00:53] yeah all that plus the cage move [00:53] our setup is soooooooo much simpler now; not quite as redundant, but the areas where we built redundancy before, in over 10 years, never needed it [00:53] our s1.lax still has over 10 years uptime [00:53] also we have spares so we can still resolve any issues quick [00:54] s1.lax uptime is 10 years, 9 weeks, 4 days, 13 hours, 10 minutes [00:54] 10 years is the standard service life of industrial electronics [00:55] it will go down, not due to failure, but because we simply don't need it anymore [00:55] I'm going to keep it in my living room for a while [00:55] hhahaah [00:55] and because it's physically huge [00:55] I'll rest my feet on it [00:55] G'night #arpnetworks thanks for your help tonight. My server is back to its normal working order. [00:55] it's also 10 years of doing a lot of changes to config [00:55] with no flash failures or anything [00:55] brycec: cool!! great to hear! [00:56] mercutio: yeah.. not even a RAM failure [00:56] s3 on the other hand [00:56] s3 is gone [00:56] s3.lax was just a software router (OpenBSD) [00:56] and yeah, s3 is dead [00:57] oh s3 was the software router [00:57] for some reason I never made that connection [00:57] yeah [00:57] iirc it was the one that always kept crashing :P [00:57] s3, then s32, then r2 [00:57] we have hardware forwarding of ipv6 now [00:57] but r2 performance was fine [00:57] acf_: b/c of Any2 IX terminating on it? That was actually s7.lax, a Cisco 6506. But it never crashed after we took Any2 off it [00:57] it's just we get to consolidate [00:57] yeah [00:57] ohh I must be thinking of s7 then [00:58] yeah it must have had some bug [00:58] t\hat was a long time ago [00:58] haha feels like yesterday [00:58] but it's done pretty well recently [00:58] s7.lax uptime is 4 years, 36 weeks, 3 days, 23 hours, 49 minutes [00:58] acf_: see how time flies? [00:58] especially when you haven't been on the irc in 3 years :P [00:59] yeah irc isn't as active as it used to be [00:59] alright guys, I'm gonna head out too... too much data center noise for today [00:59] well the +r really reduced user count [00:59] but we had to do it b/c of the spam [00:59] * up_the_irons packs up [00:59] well even before that it was pretty quiet [01:00] yeah I remember going on here for a minute during the supermicro thing [01:00] irc logs had been basically empty for days [01:00] supermicro things? [01:00] "the big hack" [01:00] oh right [01:01] seems like it's picked back up a bit the last week or so with all the stuff going on though [01:01] a little [01:01] https://danluu.com/input-lag/ [01:01] re what you were saying earlier [01:02] you've seen this yeah? [01:02] have you done the reaction speed test? [01:02] https://www.humanbenchmark.com/tests/reactiontime [01:02] what i found most interesting was that my reaction time would get slower when i was tired, but i didn't feel like my reaction time was slower [01:03] looks like I'm at roughly 300ms right now [01:03] it's kind of like how they say that driving sleep deprived can be as bad as driving drunk [01:03] i'm usally aruond 220 [01:03] i'm like 280 when tired [01:03] mouse can make a difference though [01:03] some add like 20 msec latency [01:04] thinkpad integrated trackpoint for me [01:04] 249 now [01:04] oh and screen can make a difference too [01:04] but if you use the same device and test yourself against yourself you can see how much you vary [01:05] oh also reaction times tend to be quicker when you're younger [01:05] lol I get like 500ms on my phone [01:06] yeah phones are notorious for lag [01:06] it's all accumulative also [01:06] so there's the input device latency plus the screen latency plus the os/software latency [01:07] also you can speed your reaction up slightly by turning your head the side slightly and using peripheral vision [01:07] hmm my older thinkpad I get 230ms [01:07] yeh device makes a difference [01:07] so 70msec changing device? :) [01:08] so like if you ssh'ed half way across the world you might notice a difference on your slower device vs your faster [01:08] but if you ssh'ed to a computer in the same city as you you probably won't [01:09] yeah that's a lot of latency [01:09] i think if you play first person shooter games as a kid you'll tend to have faster reaction times too. but at the same time if you have slow reaction times you probably hate first person shooters. [01:10] hahaha yeah I've never played fps [01:10] probably any twitch type games would help it [01:10] ever played vvvvvv? [01:10] or sporting things that use reaction time [01:10] nope [01:11] https://upload.wikimedia.org/wikipedia/commons/5/52/VVVVVV_-_The_Tomb_of_Mad_Carew.png [01:11] only game that comes to mind I've played that requires good reaction times [01:12] i've been playing factorio recently [01:12] which isn't really reaction time game [01:12] oh yeah old games used to be so hard! [01:12] and there would be things like you die once and you go back to the start of the game [01:13] actually that's my favourite thing about vvvvvv [01:13] that little "C" there is a checkpoint [01:13] they're quite often [01:13] ahh [01:13] you only go back to that, and there is no concept of lives [01:13] yaeh i think checkpoints were nice tbh [01:13] so you can just keep trying until you make it [01:13] otherwise I'd give up after 5 minutes :P [01:13] i think games used to be too hard [01:13] but now they're too easy instead generally [01:13] can't stand re-doing a whole level because I hit a spike [01:13] yeh [01:14] that was so common though! [01:14] I guess it instils discipline [01:14] well also less code [01:14] haha really [01:14] well things like sega master system had kilobytes of code i think? [01:14] like 20k? [01:15] yeah back in the day when video games were written in assembly [01:15] well sega soem games were bigger than others [01:15] it has 8kb ram 16kb vram [01:15] so yeah i wouldn't be surprised if games were 20k [01:16] 20k including graphics? [01:16] idk the code and graphics would probably be stored on rom right? [01:16] oh they went up to 1024kb [01:16] yeah so the roms can be quite big [01:17] surprising [01:17] yeah small ram makes sense [01:17] quite a few old amiga games didn't use the whole floppy [01:17] like pre native games [01:17] the performance vs laziness (or "work") tradeoff you mentioned earlier is pretty interesting [01:17] hmm the sega master system came out at $200 [01:18] that's cheaper than i thought [01:18] i wonder what equivilent that is now [01:18] yeah around 2001 or so languages liek perl, pyhton etc were taking off a lot [01:18] using lots of ram and having subpar performance [01:19] java is what really comes to mind [01:19] heh [01:19] yeah [01:19] when you say lots of ram and subpar performance [01:19] so java when it came out was really slow [01:19] but that would have started in the 90s yeah? [01:19] so android is java [01:19] yeah in the 90s [01:19] solaris used to be called slowaris [01:19] it was always a little sluggish [01:19] linux had a reputation for being fast and unstable [01:20] solaris for being slow and reliable [01:20] but lots of old solaris programs would fail pretty badly at times [01:20] java's awful performance seems to have kept up with improvements in computing power somehow [01:20] yeah there's been so many new versions [01:20] eclipse is still super slugish and uses all my ram [01:20] and it still seems to be kind of sluggish. also the amonut of ram programs use these days is insane [01:21] ios vs android is actually a good example of compiled vs jit code [01:21] yeah well some people thought it would be a good idea [01:21] there are some situations where ios is a lot better than android [01:21] like running a bit of code when a phone has to wake up [01:21] to make desktop applications in html / javascript and ship and entire chrome installation with them [01:21] so like if your phone is mostly idle... you can turn the cpu on for a fraction of a second and not actually warm up the caches [01:21] so you leave your caches cold to use less power [01:21] hm [01:22] so when you have compiled code it's much easier to stay within a tight cache instruction window [01:22] and that can directly transfer to better idle power consumption [01:22] although there are other tricks in use too, like wifi will do arps etc with hardware offload [01:23] afaik today Android's Java runtime actually does ahead of time compiling [01:23] oh it does doesn't it [01:23] although no idea about performance. I'm sure there is still tons of overhead from the java rutime [01:23] but yeah .. so iphones actually end up having less ram than android phones [01:23] for similar performance level [01:23] yeah I'm not surprised [01:24] mfw my phone has a higher display resolution and more ram than my laptop [01:24] heh my tablet is 2560x1600" [01:24] my phone is 2560x1440 [01:24] my laptop was 1366x768 until earlier this year [01:24] my monitors are 2560x1440s too [01:24] now 1920x1080 [01:24] my laptop is 1920x1080 too [01:24] phone is 2560xSomething [01:24] my phone is from 2014 hah [01:25] galaxy note 4 [01:25] ahh nice [01:25] it's not a bad phone [01:25] i think it has like 3gb of ram [01:25] until last year I had a Nokia windows phone from 2013 [01:25] if MS had continued supporting the platform I would have continued buying it [01:25] yeah 3gb [01:25] but oh well [01:25] my friend who works for google was saying that windows phone was good [01:26] yeah tbh I think people don't give it a chance [01:26] because desktop windows sucks, and it has the same name [01:26] it's odd, he used to (probably still does) use openbsd etc. [01:26] so then he comes and tells me windows phone is good [01:26] and i'm like hmm.. [01:26] haha yeah. I run linux on all my laptops, etc.. and (used to) windows on all my phones [01:26] apparently it's been canceled [01:26] yeah that was sad [01:26] so it's good? [01:27] imo it was very good [01:27] curious [01:27] i wonder why it never took off [01:27] i suppose lack of apps [01:27] yeah that's probably the big one [01:27] actually i know someone else who said that it was good but he's a windows fan boy [01:27] my thing is I don't like installing a ton of random apps anyway so I'm fine with it [01:27] yeh i don't ilke too many apps either [01:28] imo the Android UX used to be awful [01:28] i hardly use my phone compared to a lot of people it seems [01:28] i tried to add a contact the other day and i had to figure out where everything was :) [01:28] like I remember samsung long ago [01:28] it's odd you go down the bottmo left to contacts.. then to the top right to add [01:28] swiping through the 3 screens of apps trying to find the one I wanted [01:28] it just seems a weird ui [01:29] on WP, I'd just search the first 3 characters [01:29] like human eye goes top left to bottom right normally [01:29] so a goood way to confuse people is go bottom left to top right [01:29] for phones swyping from the left etc isn't actually that bad a way of doing things [01:29] pushing your thumb bottom left seems odd though [01:29] also you have a samsung right [01:29] yeah [01:29] so they make everything novel [01:30] iirc samsung replaces all the default android stuff with super shit samsung stuff [01:30] yeah it does [01:30] some of it's nice [01:30] last I checked (galaxy 8) they still do it [01:30] and it's still bad [01:30] but some of it's just odd [01:30] one of my biggest peeves with them [01:30] yeah i dunno some things are nice on it though [01:30] is their software all updates separately from the both the "system" and the Google store [01:30] yeah i wish they did more updates [01:30] it's like the least you could do is have a unified update mechanism [01:30] note 4 has been abandoned pretty much [01:31] all this inconsistency just feels gross [01:31] and the newer notes are kind of lame [01:31] like they took out microsd and replaceable battery [01:31] I heard note 7 is good [01:31] like i have a replacement battery on my phone [01:31] i don't have a microsd, but i like the option none the less. [01:31] i used to use a microsd with phones [01:31] but i just don't want to store lots on it anyway [01:32] yeah I just get a big internal flash and use that [01:32] microsd is kinda slow r/w performance anyway [01:32] oh wow apparently it does 24bit/192khz audio [01:34] yeah there is uhs-1 that speeds things up a bit but it's not supported on most older phones [01:34] also sdcard doesn't tend to have good wear levelling [03:02] *** r0ni has quit IRC (Read error: Connection reset by peer) [09:11] *** ziyourenxiang has quit IRC (Ping timeout: 250 seconds) [16:15] brycec: did you notice openbsd 6.5 upgrade was weirdly fast? [16:15] i think openbsd must have improved their disk performance... [16:27] i can't see anyhting in the release notes about why though [16:29] It was fast! [16:30] I always accidentally pull from the default ftp4 and have to restart from a reasonable mirror [16:31] oh i did it from mirrors.arpnetworks.com [16:31] but the pulling/installing bit seemed to be noticably less sluggish [16:33] Hacking on virtio(4), including defines, bug fixing and pci device list. [16:33] wonder if that was it [16:41] Ah nice. Yeah I've got it on a pcengines box [16:41] Home router :) [16:43] ah [16:44] is pcengines x86? [16:44] It's an amd64 clone iirc [16:45] ah ok [16:45] Apu.4 [16:45] i saw they just moved mips to clang [16:45] oh i think they moved to the clang linker too [16:45] so compiling should be faster... [16:45] Yeah [16:46] Every time I upgrade obsd something breaks work dhcpcd, which I need for ipv6 :( [16:46] odd [16:46] But the maintainer is great and super responsive [16:46] they changed openbgpd quite a lot in 6.4 [16:46] I'm probably one of like 3 people using it [16:47] it's kind of cool that they're active, but yeah can require kind of lot mangling to get things right sometimes [16:50] *** ziyourenxiang has joined #arpnetworks [17:47] has anyone else upgraded to OpenBSD on a VM yet? [18:21] up_the_irons: something seems wrong with the ARP DNS servers? My monitoring is going crazy because of timeouts and it seems to be to do with reverse DNS resolution. [18:23] Changing to 1.1.1.1 resolved. [18:30] you can tag me too mhoran ;) [18:30] reverse dns on what? [18:31] Postfix. [18:31] i can resolve 1.1.1.1 fine on both the arp dns servers [18:32] I had the ARP servers in /etc/resolv.conf and my SMTP checks were timing out. I changed resolv.conf to 1.1.1.1 and they're good again. telnet to matthoran.com 587 was taking ~5s to get a banner, which IIRC is the resolv.conf timeout before it tries another server. [18:32] weird [18:32] yeh the default is stupid high [18:32] 3 seconds is better now days [19:13] *** hive-mind has quit IRC (Ping timeout: 250 seconds) [19:21] *** hive-mind has joined #arpnetworks