| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Comments:
<0> anyone got any idea how long reiserfsck will take to check a 4 terabyte fairly full volume
<1> long
<2> ya lol
<0> as in hours?
<1> easily
<1> although it heavily depends on CPU and disk speed of course
<3> SymmDC, go to sleep now.. when it's done you'll be waking up.
<0> oh **** this, I aint gonna run the fsck then
<1> it shouldn't be necessary normally
<1> it's a journaling fs
<3> what seems to be the problem pinguin ?
<1> don't ask me :)
<3> i mean SymmDC
<0> heh tsu just the 4th time that box has had to be hard powered
<0> was getting a little worried about it considering that I manually mount those fs's every time
<0> /dev/sdb1 1.1T 700G 418G 63% /mirror/internal
<3> SymmDC, hardware is failing.. have you checked your memory?
<0> 3.9T 1.3T 2.7T 33% /mirror/iscsi-san-1
<0> 3.9T 1.2T 2.8T 30% /mirror/iscsi-san-2
<0> 3.9T 2.2T 1.8T 56% /mirror/iscsi-san-3
<3> hard locks are few and far between
<0> tsu nah, hardware isnt failing, its a memory leak in 2.6.16 regarding gigE ethernet cards and adjusted tcp window sizes
<0> that much I know for a fact
<1> I've never experienced a hardlock
<0> unfortunatly Im stuck with 2.6.16 because of the iscsi support
<1> and 2.6.17 ?
<0> same deal
<0> Im switching to some iSCSI HBA cards tomorrow
<0> which will at least deal with the gigE load there
<1> O_o
<0> that will at least seriously slow down the memory leakage
<1> wait
<1> did you say memory leaks with gigabit ethernet ?
<0> at high speed with adjusted tcp window sizes yes
<1> how come I don't have trouble with that :X
<0> the memory leaks badly, particularly on intel e1000's
<0> pinguin you run any of your boxes with adjusted window sizes at 200+mbit/second for 2 hours at a time?
<1> I did when I reinstalled my storage box
<1> took hours
<1> to move the data back
<0> what did you adjust the window sizes 2
<1> are you sure the problem is not purely related to the window sizes?
<1> and how come this hasn't been solved if it's such a serious issue ?
<4> mistik1: you around?
<0> pinguin look at the netdev list
<0> it was talked about on there
<0> and there was a patch given there as well, that patch doesnt work
<4> nastjuid: yea, I'm enjoying my priest too much.. I didn't DE everything (kept the staff, one of my rings, and my cloak) buth otherwise, if it was DEable it was DEd
<0> pinguin its *VERY* definately there, because you hit a point where you start getting a ton of kernel debugging error messages and it refuses to allocate swap because its lost track of memory because of leaks
<1> I had errors a while ago with earlier 2.6.x kernels with adjusted window sizes
<1> tcp_rmem etc.
<1> they're gone now that I use 2.6.17
<1> or just haven't appeared yet maybe
<0> the other problem with 2.6.17 is that it breaks iscsi
<0> breaks it badly :p
<0> Im hoping to be able to use a qlogic HBA in here after tomorrow and then take the box to 2.6.17 again and try again
<1> lemme know how thta works out
<1> :)
<5> kitchen: i wish i could pick a gnome up and hit it with the DE wand thingy
<5> zaps gnomes dead
<6> I swear people in this company seem to enjoy making things more difficult for the guys on the "lower end"
<6> we call up ****ing 210 cusotmers to tell them their static IP is chaning and the engineers say 'woops sorry, it's the 10th now'.
<0> pinguin heh just hope I dont have **** getting those HBA's installed
<0> because they use commercial drivers rather than opensource ones
<0> I dont think there are any qlogic HBA drives available
<1> nastjuid: this is normal behavior. Please remain seated until a specially-appointed colleague comes to pick you up and escort you to the special learning room.
<0> and those commerical ones were designed specifically for redhat, but I should be able to hack it together to work with gentoo
<4> bitrot: ouch.
<6> kitchen: it's bull**** man.
<6> If we have a "Network Blackout"
<1> bitrot even
<6> our engineers should know of this
<6> The hoarding of information around here is disgusting, seriously.
<4> nastjuid: i should be rank 4 tomorrow.
<4> won like 200 ab games this weekend
<6> Everyone is so afraid of losing their ****ing job that they horde information.
<6> Just to casue more issues, that's what I see.
<1> that's normal everywhere
<6> Yeah I know
<6> it makes me sick
<6> people are ****ing stupid
<4> yea, it's called job security
<6> no
<6> That's not job security
<4> no i mean
<6> being GOOD at your job is job security
<6> Hording information is calling being a douche bag.
<4> the person is hoarding the information to ensure that he remains with the company.
<0> heh bitrot dont worry, I have huge issues with my provider
<0> though admittedly I have an advantage in dealing with them that most people dont have to get things sorted out
<6> It's saying I feel that I'm not valubale in any way other than having these little pieces of information that I don't leak to anyone, I simply sit on it like a moron so I feel powerful.
<0> heh, my provider tends not to piss me off 2 badly
<1> bitrot
<6> pinguin:
<1> if that was true, I wouldn't have been fired from my last provider
<1> eh
<6> pinguin: refer back to people are ****ing stupid.
<1> last two employers
<0> bitrot unfortunatly the only way to really convince a provider to take you seriously
<0> in my experience
<1> they fired me twice because I became a risk. In no covert terms I became a threat to THEIR job security
<0> is to spend so much money with them that they cant afford to **** you around
<6> SymmDC: I AM THE PROVIDER
<0> heh bitrot so who was screwing you around, the customers?
<7> Drug dealing/
<7> ?
<8> is there a command to tell me what model my video card is?
<8> dmesg grep something?
<9> it sounds like the upper level techs planned an outage or IP change, then bitrot and his crew had to call the customers, when that was done, the upper level tech changed the dates or something
<0> :( unfortunatly with some of my clients spending a million dollars a year I cant afford to NOT take their crap
<7> tendo: lspci
<0> duhbLow7 heh, yeah that can ****
<0> I dont have that problem, because I am the upper tech levels hahaha
<9> haha, yeah, i'm the douche bag with all the secrets
<9> :D
<0> duhbLow7 Im the CTO of my company
<0> :p
<9> nice
<0> nothing happens technically without my say so, and if I say no, or wait, or hold off, by god someones head will roll if instructions arent followed to the letter
<9> i'm the highest level IT person that knows what the hell our servers do
<9> my boss and his boss are more of project management
<9> they tell me what to build
<0> duhbLow7 heh, I wont even let my subordinates ONTO half my servers
<0> never mind my routers
<0> the p***words and the rest are escrowed into a safe incase I drop dead
<7> i'm the highest level person in my company that knows what the hell the company actually does
<8> how can i display what video card driver the system is currently using?
<0> but while Im alive no one touches that ****
<8> i think i must have an old one or something
<0> heh, its 2 much of a financial risk to the company to have people playing on my equipment
<7> true
<9> must be nice
<7> it's a two person company and the other person is an accountant
<7> but still....
<0> heh einride hahahaha die
<10> SymmDC: it's kind of funny when you announce * from one uplink to the other uplink.
<0> einride hahahaha I made that mistake on my v6 before
<10> suddenly you route a freaking country on 155 + 45 mbit
<0> and it ended up with me transitting half the world to occaid
<0> and vice versa
<0> :p
<1> baaha
<10> yeah that kind of whoops
<1> that's neat
<0> einride we also had a situation where *cough* we accidently leaked the entire national routing table to geant
<10> haha
<0> which propogated all the way through to abeline, geant and tein2
<0> and aarnet
Return to
#gentoo
or
Go to some related
logs:
jewlex
#heroin
#nhl
#debian
slap happy deluxepass
sata sda* not valid block device
#computers
the ducks scott, jono
#windows
#visualbasic
|
|