slow network

Philip Greenspun's Homepage : Philip Greenspun's Homepage Discussion Forums : 6916 : One Thread
Notify me of new responses
this is ridiculous! it's impossible to get work done. the network hang minutes at a time and it's just not possible to get any work done. the network seems to be fine:
sun/: traceroute lcsweb93.lcs
traceroute: Warning: ckecksums disabled
traceroute to lcsweb93.lcs.MIT.EDU (18.43.3.93), 30 hops max, 40 byte
packets
 1  NW12-RTR-PHI-BETA-EPSILON-ETHER.MIT.EDU (18.236.0.1)  1.124 ms 
3.116 ms  1.181 ms
 2  B24-RTR-FDDI.MIT.EDU (18.168.0.14)  3.110 ms  1.217 ms  1.125 ms
 3  RADOLE.LCS.MIT.EDU (18.201.1.3)  3.737 ms  1.328 ms  1.326 ms
 4  anacreon.lcs.mit.edu (18.24.10.1)  1.540 ms  1.286 ms  1.371 ms
 5  webwork3.lcs.mit.edu (18.43.3.83)  2.008 ms  1.457 ms  1.827 ms
the load seems to be fine:
last pid: 16383;  load averages:  1.81,  2.09,  2.10                  
05:33:49
127 processes: 109 sleeping, 17 running, 1 on cpu
CPU states:  0.0% idle, 97.4% user,  2.6% kernel,  0.0% iowait,  0.0%
swap
Memory: 512M real, 8096K free, 1470M swap in use, 67M swap free

  PID USERNAME THR PRI NICE  SIZE   RES STATE   TIME    CPU COMMAND
13826 nsadmin   12 -25    0  991M   18M run   107:33 94.08% nsd82
16383 std93      1  30    0 1496K 1312K cpu     0:00  0.85% top
13821 nsadmin    9  35    0   25M   10M sleep   1:23  0.31% nsd82
13818 nsadmin    9  35    0   24M   10M sleep   1:21  0.31% nsd82
13839 nsadmin    9  34    0   23M 9952K sleep   1:23  0.22% nsd82
  878 ora8i      1  23    0  379M  344M sleep   0:35  0.11% oracle
13844 nsadmin    9  35    0   23M 9912K sleep   1:21  0.09% nsd82
13845 nsadmin    9  34    0   25M   10M sleep   1:26  0.09% nsd82
13805 nsadmin    9  35    0   22M   10M sleep   1:22  0.09% nsd82
  874 ora8i      1  24    0  379M  345M sleep   0:35  0.07% oracle
 3838 root       1  33    0 1896K  336K sleep   0:23  0.04% sshd1
  522 root       1  34    0 1568K  344K sleep   0:01  0.04% ttymon
 3849 std90      1  34    0 9272K  664K sleep   1:03  0.03% emacs
14969 root       1  33    0 1896K  424K sleep   0:10  0.03% sshd1
14972 std93      1  23    0 2024K  728K sleep   0:00  0.03% bash
whatever the problem is, it's intermittent and very annoying. please fix it, or at least find some solution. perhaps putting servers on the main mit network instead of the lcs network? or even just telling me how to set up all the oracle, aolserver, acs stuff on my own linux box would make me happy... i'm really sick of wasting hours and hours because i can only type 2 lines every 5 minutes

-- Tony Chao, March 9, 2000

Answers

Student server 186 was running amuck (you can see in the top message that the webserver was consuming a lot(!) of resource). I restarted it.

We'll distribute linux 'team' accounts to interested parties.

-- James Buszard-Welcher, March 9, 2000