up_the_irons: try it on 2.5 < x < 3 otherwise weird things happen nlayer is droppin' mah packets Packets Pings Host Loss% Snt Last Avg Best Wrst StDev 8. ae2-40g.cr2.ord1.us.nlayer.net 13.3% 30 19.7 19.7 17.0 28.8 3.1 12. xe-4-2-1.cr1.lax1.us.nlayer.net 3.6% 29 80.0 79.5 75.6 115.9 7.6 13. ae1-50g.ar1.lax2.us.nlayer.net 50.0% 29 80.6 80.3 77.8 85.7 2.5 everypony Hello guys. I need your suggestions; What online creditcard payment merchant would you suggest for a tiny online shop? I'm looking for lowest-possible membership/transaction cost. Paypal takes a lot of money, as well as 2checkout. So any idea? Swipe and Square are popular (I think those are the names... not positive, as I don't use either) s/Swipe/Stripe Stripe seems good, thanks brycec. But it's Turkey, not the U.S. Well.. Yeah, I can't help you with Turkey, sorry Many thanks anyway. google wallet deXar: compared to a bank merchant account, paypal does not take a lot. banks have tons of hidden fees. we use paypal merchant services. up_the_irons: wait my drives are hotswap?? hazardous: wut i just had the realisation that my server never rebooted from days ago hazardous: yeah of course they are hot swap ;) damn i'm slow on the plus side i finished finals last week so i can actually sleep for once instead of wandering around in a hazy zombielike state hah cool hazardous: did your raid rebuild on its own? still rebuilding 56% or so now when did you start it on the rebuild? immediately after you replaced it i have no idea why it's taking this long personally oh damn yeah that seems slow it was like 0% for hours so i was really confused i should check smartmon or something or see what kind of io i get you might have it on the slow rebuild setting (keeping io fast) What level raid? 1 brycec: mdadm sw 1 Yeah... that should've been done days ago mdadm default raid1 speed is 20MB/s IIRC Filesystem Size Used Avail Use% Mounted on /dev/md0p1 899G 62G 793G 8% / im not even using that much disk so :V Well mdadm is fs-agnostic State : clean, degraded, recovering Rebuild Status : 63% complete it's still block for block, doesn't matter what you use [============>........] recovery = 63.9% (613462784/960010104) finish=5593.9min speed=1032K/sec 1032 ..? ouch, that's slow 900GB @ 20MB/s = 41,890 seconds which is about 12hours hazardous: sysctl dev.raid.speed_limit_min sysctl dev.raid.speed_limit_max Those are your "speed limits" dev.raid.speed_limit_min = 100000 wtf? and my max is 200k That's MB the units on that are KB *kB hazardous: Try fiddling with the bitmap? http://www.cyberciti.biz/tips/linux-raid-increase-resync-rebuild-speed.html mdadm: Cannot add bitmap while array is resyncing or reshaping etc. mdadm: failed to set internal bitmap. ha damn this might be really random but what's the sanest way to get around la/orange county without a car I'm guessing "get a car" figures last time i was in irvine i found out buses stopped running at noon and started again at 4pm boy was that fun (but I'm not in Cali... only passed through there a handful of times) i'm from the part of cali where it's easier/faster to just take public transit because driving is gridlock 24/7 unfortunately this does't seem to be a thing in socal :< i take the subway up_the_irons: are you on ipv6 any2ix? yeah ok cool someone probably decided peering with he.net ipv6 was good idea by someone it seems i mean reach and the reason i couldn't google that was because of alternative syntax for ipv6 addresses :: verus :0:0:0:0: it seems ntt has been stable but lots of routes still go over nlayer is the plan to shift another provider to that network soon then default route as long as it's up? until shifting everything right now all outgoing is NTT, cuz i have a full routing table and i'm also testing outbound on it. long term plan is to get another session with GTT going (basically, nLayer now), full table, mix it with NTT (also full table), and then also distribute all my peering routes in that mix, so the outbound will be really killer oh it changed oh is that only for people terminating on that router though? i wish nlayer would stop deprioritising icmp so much like that guy showing the "packet loss" up above but nother than that mzima/nlayer/gtt seem good oh it did change, this must have been recentish :) it still seems peering etc goes outbound normally ie via peering but default did seem to change is it just ipv4 doing that? or is ipv6 like that too? he.net seems very popular for ipv6, but dunno if that's just cos lots of people have ipv6 peering with he.net like undeadly.org is ntt with ipv4, he.net with ipv6 Isn't HE's IPv6 peering free? for tunnels that gives weird mtu's though :/ well he.net have open peering policy anyway i think but that's the same for ipv4/ipv6 http://www.he.net/peering.html they don't say anyhting about traffic ratios or anything or minimum bandwidth utilisation i am trying to find ipv6 looking glass on any2ix packet clearing house have an ipv4 one mercutio: In the UK at least, and I think it's the same anywhere, HE will give free IPv6 transit to anyone who is at a peering point they are also at mercutio: yeah, peering traffic is still going to peers. but otherwise, goes to NTT atm any staff around? you've got deployments! :p cc: up_the_irons sweet, thanks. I wouldn't care, but I want to get some work done. sorressean: i think i replied to your ticket; asking for account info or wait... that might have been someone else sorressean: ok, i see your order now. sorressean: i have a meeting at 2:15, but i think i can squeeze out that vm... up_the_irons: sweet, thanks! sorry for the late reply. :) no worries sweet! :) plett: ahh transit rather than peering mercutio: The difference is jut a matter of some BGP export filters *just lol "squeeze out that vm"