today i had a lot of email from cron with an error message from my
script that check my internet connection. the message was:

ping: sendto: No buffer space available

i had a similar error from named about in the same time interval. so i
googled a bit but all i found was about some possible DoS.

i also found that one of my pptp processes died (even if i'm not sure
when it died) so can i suppose i really had a DoS? (i was not at home
when it might have happened).

anyway since today i had again the problem i described a couple of days
ago (my lan not working) i also checked some things using the paper
"running and tuning openbsd etc" as a reference (btw it seems the lan
problem is just the ethernet cable or the nic connector giving some
trouble)

this is what i found:

$ netstat -f inet
Active Internet connections
Proto Recv-Q Send-Q Local Address Foreign Address (state)
tcp 0 0 meshuggah.ssh opeth.1058 ESTABLISHED
tcp 0 0 cerbero.19568 caronte.pptp ESTABLISHED
tcp 0 0 meshuggah.8080 *.* LISTEN
tcp 0 0 meshuggah.nut meshuggah.17449 ESTABLISHED
tcp 0 0 meshuggah.17449 meshuggah.nut ESTABLISHED
Active Internet connections
Proto Recv-Q Send-Q Local Address Foreign Address (state)
udp 0 0 localhost.48661 localhost.48661
udp 0 0 meshuggah.netbios- *.*
udp 0 0 meshuggah.netbios- *.*
udp 0 0 cerbero.domain *.*
udp 0 0 meshuggah.domain *.*
udp 0 0 localhost.domain *.*
$ netstat -m
143 mbufs in use:
        130 mbufs allocated to data
        5 mbufs allocated to packet headers
        8 mbufs allocated to socket names and addresses
130/162 mapped pages in use
420 Kbytes allocated to network (70% in use)
0 requests for memory denied
0 requests for memory delayed
0 calls to protocol drain routines

$ vmstat
 procs memory page disks faults cpu
 r b w avm fre flt re pi po fr sr wd0 wd1 in sy cs us sy id
 010 0 41308 32416 11 0 0 0 0 0 1 0 281 278 79 0 1 99

if i understood what i read in the paper it seems there are some
processes slowed by too many interrupts, and given the fact that i have
no queue on the internet side, i suppose there's something going wrong
on the lan nic, and since the problem i had with my lan was, as i said,
either the cable or the nic connector, can i assume that the nic on the
lan side is probably faulty? (it's an intel etherexpress btw, but is
also quite old)

TIA

-- 
    Shiva
  "Better true to yourself
Than a perfect shadow
       Of somebody else
     An empty shell"
(MrBig, My new religion)

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]