PDA

View Full Version : Issues ?!?



MDFaunce
11-25-2002, 09:32 PM
I noticed a couple of things.

While running SB client on Windows, it keeps some TCP connections intact:

C:\WINDOWS>netstat | find "sb.pns.net"
TCP shade:1315 sb.pns.net:1717 CLOSE_WAIT
TCP shade:1367 sb.pns.net:1717 CLOSE_WAIT
TCP shade:1420 sb.pns.net:1717 CLOSE_WAIT
TCP shade:1459 sb.pns.net:1717 CLOSE_WAIT
TCP shade:1496 sb.pns.net:1717 CLOSE_WAIT
TCP shade:1497 sb.pns.net:1717 CLOSE_WAIT
TCP shade:1500 sb.pns.net:1717 CLOSE_WAIT
TCP shade:1504 sb.pns.net:1717 CLOSE_WAIT

If left around long, enough, you get more and more of these. It looks to me like these connections are only closed when you submit a completed n or stop/start the client.

Also, I've gotten some strange displays on some blocks. Basically, it looks like when the n is very small that messes up some internal computations. It says it's going to complete the full test *before* it completes the block, even though there is only one block. Also the time remaining is messed up when this happens. I never noticed it on the higher n runs, but on these recent ones with small n it's been a lot more frequent.

shifted
11-25-2002, 11:15 PM
Originally posted by MDFaunce
I noticed a couple of things.

While running SB client on Windows, it keeps some TCP connections intact:

C:\WINDOWS>netstat | find "sb.pns.net"
TCP shade:1315 sb.pns.net:1717 CLOSE_WAIT
TCP shade:1367 sb.pns.net:1717 CLOSE_WAIT
TCP shade:1420 sb.pns.net:1717 CLOSE_WAIT
TCP shade:1459 sb.pns.net:1717 CLOSE_WAIT
TCP shade:1496 sb.pns.net:1717 CLOSE_WAIT
TCP shade:1497 sb.pns.net:1717 CLOSE_WAIT
TCP shade:1500 sb.pns.net:1717 CLOSE_WAIT
TCP shade:1504 sb.pns.net:1717 CLOSE_WAIT

If left around long, enough, you get more and more of these. It looks to me like these connections are only closed when you submit a completed n or stop/start the client.

I have about 80 of these open right now. CLOSE_WAIT's usually linger around for a few minutes... but not days :/ Probably a small bug in the code.


Also, I've gotten some strange displays on some blocks. Basically, it looks like when the n is very small that messes up some internal computations. It says it's going to complete the full test *before* it completes the block, even though there is only one block. Also the time remaining is messed up when this happens. I never noticed it on the higher n runs, but on these recent ones with small n it's been a lot more frequent.

Yep. There are some calculation details being worked out, no doubt, by Louie, for both the win client and the stats online.

jjjjL
11-25-2002, 11:40 PM
MDFaunce - i already wrote you an email about this so i don't understand the post. there is certainly no "[messing] up some internal computations". the client is behaving fine.

yes, it is true that in the last block, the block compeletion time isn't correct but that is a small point. usually isn't noticed although the low n's make it more noticable. for now, ignore that and i'll possibly change it in the future.

that's an interesting observation with windows networking. i'd buy that there are 8 open connections. "about 80" as shifted claims sounds a bit high. personally i have 12 on my workstation here.

blame Gates. ;) basically i don't reuse connections. i close them after each transmission rather than keeping a constant connect. so it opens a new socket for each transmission. my server closes it but windows must not believe it. netstat obviously works in linux so if there is a similar "problem", i'll look into doing something about it.


-Louie

shifted
11-26-2002, 01:44 AM
Originally posted by jjjjL
that's an interesting observation with windows networking. i'd buy that there are 8 open connections. "about 80" as shifted claims sounds a bit high. personally i have 12 on my workstation here.

Actually, it's 130 right now (client has been running for over 107 hours.) You can count them, if you wish ;) However, i haven't noticed any ill effects, so i'm not concerned.


C:\>netstat | find "sb.pns.net"
TCP shiftedbox:3019 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3021 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3049 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3055 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3075 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3076 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3089 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3098 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3103 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3111 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3122 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3160 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3162 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3181 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3197 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3210 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3211 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3225 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3226 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3232 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3266 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3285 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3294 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3297 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3308 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3320 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3341 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3353 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3393 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3398 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3400 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3406 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3458 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3469 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3491 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3504 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3509 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3528 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3542 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3544 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3547 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3562 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3564 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3581 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3582 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3606 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3607 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3629 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3647 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3648 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3670 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3681 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3708 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3726 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3752 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3791 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3799 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3807 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3812 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3862 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3868 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3880 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3884 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3909 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3912 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3915 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3938 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3939 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3959 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:3960 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4006 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4018 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4036 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4056 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4078 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4102 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4119 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4140 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4149 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4152 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4187 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4208 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4219 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4232 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4243 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4267 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4275 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4317 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4330 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4368 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4369 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4375 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4382 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4398 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4431 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4487 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4525 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4539 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4561 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4578 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4608 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4609 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4618 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4636 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4645 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4664 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4674 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4681 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4692 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4716 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4717 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4735 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4763 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4772 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4793 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4801 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4807 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4820 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4841 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4842 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4870 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4875 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4886 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4908 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4923 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4943 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4961 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4965 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4983 sb.pns.net:1717 CLOSE_WAIT
TCP shiftedbox:4985 sb.pns.net:1717 CLOSE_WAIT

C:\>

smh
11-26-2002, 03:09 AM
yes, it is true that in the last block, the block compeletion time isn't correct but that is a small point. usually isn't noticed although the low n's make it more noticable. for now, ignore that and i'll possibly change it in the future.

There are more strange things with the time remaining. Sometimes it's a very high (tens of thousands hours), sometimes it counts down to zero, and sometimes the time remaining is the time the client is running plus a couple of seconds.

No big deal though, just thought i mention it.