Network Performance Problems?

Network Performance Problems?

Post by Frank Parkinso » Fri, 11 Aug 2000 04:00:00



Our users start to complain about slow responses over a WAN and LAN when
the network loading as shown by "monitor" under "en0" exceeds about
30kB/s.
The network card is an IBM 10/100 mbps auto sensing PCI adapter.

The system is a quad processor S7A running AIX 4.3.3; the monitor output
is shown here.

Direct connect users (connected by SIO) show no deterioration of
responses.

Any ideas? Thanks in advance.

AIX System monitor v2.1.7PRE7 21may2000: w+p2          Thu Aug 10
14:44:19 2000
Uptime:  11 days, 03:12   Users: 203 of 524 active 246 remote 35:00
sleep time
CPU: User 41.5% Sys 40.4% Wait 11.8% Idle  6.3%   Refresh:  1.68 s
0%             25%              50%               75%              100%

Quote:>>>>>>>>>>>>>>>>>>>>>>>>>>>>>=============================--------

Runnable (Swap-in) processes 14.91 (2.98)  load average:  3.60,  4.93,
5.89
Memory    Real     Virtual    Paging (4kB)    Process events
File/TTY-IO
free     6192 MB      0 MB    5281.5 pgfaults    1244 pswitch       2
iget
procs  4188075 MB   1408 MB      1.8 pgin       67533 syscall   11740
namei
files    6180 MB                98.4 pgout      20400 read       3821
dirblk
total    6144 MB   1408 MB       0.0 pgsin       3450 write
13552856readch
IO (kB/s) read  write busy%      0.0 pgsout        64 fork     440928
writech
hdisk0     0.0     0.0    0                         60 exec        106
ttyrawch
hdisk1     0.0     0.0    0   Client Server NFS/s    0 rcvint        0
ttycanch
hdisk2     2.4     2.7    0      0.0    0.0 calls    0 xmtint      211
ttyoutch
hdisk3     0.0     2.7    0      0.0    0.0 retry    0 mdmint
hdisk4     0.0   47.7    0      0.0    0.0 getattr
hdisk5     0.0   12.5    0      0.0    0.0 lookup       Netw   read
write
kB/s
hdisk6     0.0     0.0    0      0.0    0.0 read        lo0     0.0
0.0
hdisk7     0.0     5.4    0      0.0    0.0 write       en0     7.2
37.1
hdisk8     0.0   136.2    0      0.0    0.0 other       en1     0.0
0.0
hdisk9     2.4    0.0     0

  fparkinson.vcf
< 1K Download
 
 
 

Network Performance Problems?

Post by Devon Caine » Fri, 11 Aug 2000 04:00:00


Have you tried other available tools to try to isolate if this is a
network problem?
ping (check for round trip times from the box to elsewhere)
traceroute (if round trip times are slow .. to see where the bottleneck
is)
Once you determine that it is the machines problem then other AIX tool
are available. netstat -m :- to see if you have any mbuf errors, if so
increase thewall value with the "no command".
Lots of other areas but just giving you somewhere to start. :)

Devon_C


> Our users start to complain about slow responses over a WAN and LAN when
> the network loading as shown by "monitor" under "en0" exceeds about
> 30kB/s.
> The network card is an IBM 10/100 mbps auto sensing PCI adapter.

> The system is a quad processor S7A running AIX 4.3.3; the monitor output
> is shown here.

> Direct connect users (connected by SIO) show no deterioration of
> responses.

> Any ideas? Thanks in advance.

> AIX System monitor v2.1.7PRE7 21may2000: w+p2          Thu Aug 10
> 14:44:19 2000
> Uptime:  11 days, 03:12   Users: 203 of 524 active 246 remote 35:00
> sleep time
> CPU: User 41.5% Sys 40.4% Wait 11.8% Idle  6.3%   Refresh:  1.68 s
> 0%             25%              50%               75%              100%
> >>>>>>>>>>>>>>>>>>>>>>>>>>>>>=============================--------
> Runnable (Swap-in) processes 14.91 (2.98)  load average:  3.60,  4.93,
> 5.89
> Memory    Real     Virtual    Paging (4kB)    Process events
> File/TTY-IO
> free     6192 MB      0 MB    5281.5 pgfaults    1244 pswitch       2
> iget
> procs  4188075 MB   1408 MB      1.8 pgin       67533 syscall   11740
> namei
> files    6180 MB                98.4 pgout      20400 read       3821
> dirblk
> total    6144 MB   1408 MB       0.0 pgsin       3450 write
> 13552856readch
> IO (kB/s) read  write busy%      0.0 pgsout        64 fork     440928
> writech
> hdisk0     0.0     0.0    0                         60 exec        106
> ttyrawch
> hdisk1     0.0     0.0    0   Client Server NFS/s    0 rcvint        0
> ttycanch
> hdisk2     2.4     2.7    0      0.0    0.0 calls    0 xmtint      211
> ttyoutch
> hdisk3     0.0     2.7    0      0.0    0.0 retry    0 mdmint
> hdisk4     0.0   47.7    0      0.0    0.0 getattr
> hdisk5     0.0   12.5    0      0.0    0.0 lookup       Netw   read
> write
> kB/s
> hdisk6     0.0     0.0    0      0.0    0.0 read        lo0     0.0
> 0.0
> hdisk7     0.0     5.4    0      0.0    0.0 write       en0     7.2
> 37.1
> hdisk8     0.0   136.2    0      0.0    0.0 other       en1     0.0
> 0.0
> hdisk9     2.4    0.0     0


 
 
 

Network Performance Problems?

Post by Mauro Marzorat » Tue, 15 Aug 2000 04:00:00


Frank:

couple of things to check above and beyond the obvious ping/traceroute.

.- enstat ent0 several times over time to see performance of the local
ethernet segment (collisions, etc), and the media speed (jabber occurring?)

.- netstat -nI en0 to see if your packets are being dropped

.- errpt -dH to see if there are any HW errors reported

.- netstat -s to see the statuses of the different protocols

Regards,
/mauro


> Our users start to complain about slow responses over a WAN and LAN when
> the network loading as shown by "monitor" under "en0" exceeds about
> 30kB/s.
> The network card is an IBM 10/100 mbps auto sensing PCI adapter.

> The system is a quad processor S7A running AIX 4.3.3; the monitor output
> is shown here.

> Direct connect users (connected by SIO) show no deterioration of
> responses.

> Any ideas? Thanks in advance.

> AIX System monitor v2.1.7PRE7 21may2000: w+p2          Thu Aug 10
> 14:44:19 2000
> Uptime:  11 days, 03:12   Users: 203 of 524 active 246 remote 35:00
> sleep time
> CPU: User 41.5% Sys 40.4% Wait 11.8% Idle  6.3%   Refresh:  1.68 s
> 0%             25%              50%               75%              100%
> >>>>>>>>>>>>>>>>>>>>>>>>>>>>>=============================--------
> Runnable (Swap-in) processes 14.91 (2.98)  load average:  3.60,  4.93,
> 5.89
> Memory    Real     Virtual    Paging (4kB)    Process events
> File/TTY-IO
> free     6192 MB      0 MB    5281.5 pgfaults    1244 pswitch       2
> iget
> procs  4188075 MB   1408 MB      1.8 pgin       67533 syscall   11740
> namei
> files    6180 MB                98.4 pgout      20400 read       3821
> dirblk
> total    6144 MB   1408 MB       0.0 pgsin       3450 write
> 13552856readch
> IO (kB/s) read  write busy%      0.0 pgsout        64 fork     440928
> writech
> hdisk0     0.0     0.0    0                         60 exec        106
> ttyrawch
> hdisk1     0.0     0.0    0   Client Server NFS/s    0 rcvint        0
> ttycanch
> hdisk2     2.4     2.7    0      0.0    0.0 calls    0 xmtint      211
> ttyoutch
> hdisk3     0.0     2.7    0      0.0    0.0 retry    0 mdmint
> hdisk4     0.0   47.7    0      0.0    0.0 getattr
> hdisk5     0.0   12.5    0      0.0    0.0 lookup       Netw   read
> write
> kB/s
> hdisk6     0.0     0.0    0      0.0    0.0 read        lo0     0.0
> 0.0
> hdisk7     0.0     5.4    0      0.0    0.0 write       en0     7.2
> 37.1
> hdisk8     0.0   136.2    0      0.0    0.0 other       en1     0.0
> 0.0
> hdisk9     2.4    0.0     0

 
 
 

Network Performance Problems?

Post by Urban A. Haa » Fri, 18 Aug 2000 04:00:00


Frank,

I won't say this IS it, but I have read in other posts that monitor under
AIX 4.3.3 is known to give bad performance statistics.

U


> Our users start to complain about slow responses over a WAN and LAN when
> the network loading as shown by "monitor" under "en0" exceeds about
> 30kB/s.
> The network card is an IBM 10/100 mbps auto sensing PCI adapter.

> The system is a quad processor S7A running AIX 4.3.3; the monitor output
> is shown here.

> Direct connect users (connected by SIO) show no deterioration of
> responses.

> Any ideas? Thanks in advance.

> AIX System monitor v2.1.7PRE7 21may2000: w+p2          Thu Aug 10
> 14:44:19 2000
> Uptime:  11 days, 03:12   Users: 203 of 524 active 246 remote 35:00
> sleep time
> CPU: User 41.5% Sys 40.4% Wait 11.8% Idle  6.3%   Refresh:  1.68 s
> 0%             25%              50%               75%              100%
> >>>>>>>>>>>>>>>>>>>>>>>>>>>>>=============================--------
> Runnable (Swap-in) processes 14.91 (2.98)  load average:  3.60,  4.93,
> 5.89
> Memory    Real     Virtual    Paging (4kB)    Process events
> File/TTY-IO
> free     6192 MB      0 MB    5281.5 pgfaults    1244 pswitch       2
> iget
> procs  4188075 MB   1408 MB      1.8 pgin       67533 syscall   11740
> namei
> files    6180 MB                98.4 pgout      20400 read       3821
> dirblk
> total    6144 MB   1408 MB       0.0 pgsin       3450 write
> 13552856readch
> IO (kB/s) read  write busy%      0.0 pgsout        64 fork     440928
> writech
> hdisk0     0.0     0.0    0                         60 exec        106
> ttyrawch
> hdisk1     0.0     0.0    0   Client Server NFS/s    0 rcvint        0
> ttycanch
> hdisk2     2.4     2.7    0      0.0    0.0 calls    0 xmtint      211
> ttyoutch
> hdisk3     0.0     2.7    0      0.0    0.0 retry    0 mdmint
> hdisk4     0.0   47.7    0      0.0    0.0 getattr
> hdisk5     0.0   12.5    0      0.0    0.0 lookup       Netw   read
> write
> kB/s
> hdisk6     0.0     0.0    0      0.0    0.0 read        lo0     0.0
> 0.0
> hdisk7     0.0     5.4    0      0.0    0.0 write       en0     7.2
> 37.1
> hdisk8     0.0   136.2    0      0.0    0.0 other       en1     0.0
> 0.0
> hdisk9     2.4    0.0     0

--
Urban A. Haas
CEO - Urban Technology, Inc.
Minneapolis, MN  USA
Phone: (952) 595-8810    Fax: (952) 595-8710

Web: http://www.urbantechnology.com

This e-mail was composed of 100% recycled bits.

 
 
 

Network Performance Problems?

Post by Urban A. Haa » Fri, 18 Aug 2000 04:00:00


Frank,

Two other ideas:
1. Check that the ent0 adapter is using maximum buffer sizes (smit, devices,
communication, ethernet, change/show) and adjust the buffers to their
maximum sizes.
2. Use no to tune the rfc1323, tcp_sendspace, recvspace and udp as well (all
4 for them) to 512K (512*1024). This can improve things a bit.

Urban


> Our users start to complain about slow responses over a WAN and LAN when
> the network loading as shown by "monitor" under "en0" exceeds about
> 30kB/s.
> The network card is an IBM 10/100 mbps auto sensing PCI adapter.

> The system is a quad processor S7A running AIX 4.3.3; the monitor output
> is shown here.

> Direct connect users (connected by SIO) show no deterioration of
> responses.

> Any ideas? Thanks in advance.

> AIX System monitor v2.1.7PRE7 21may2000: w+p2          Thu Aug 10
> 14:44:19 2000
> Uptime:  11 days, 03:12   Users: 203 of 524 active 246 remote 35:00
> sleep time
> CPU: User 41.5% Sys 40.4% Wait 11.8% Idle  6.3%   Refresh:  1.68 s
> 0%             25%              50%               75%              100%
> >>>>>>>>>>>>>>>>>>>>>>>>>>>>>=============================--------
> Runnable (Swap-in) processes 14.91 (2.98)  load average:  3.60,  4.93,
> 5.89
> Memory    Real     Virtual    Paging (4kB)    Process events
> File/TTY-IO
> free     6192 MB      0 MB    5281.5 pgfaults    1244 pswitch       2
> iget
> procs  4188075 MB   1408 MB      1.8 pgin       67533 syscall   11740
> namei
> files    6180 MB                98.4 pgout      20400 read       3821
> dirblk
> total    6144 MB   1408 MB       0.0 pgsin       3450 write
> 13552856readch
> IO (kB/s) read  write busy%      0.0 pgsout        64 fork     440928
> writech
> hdisk0     0.0     0.0    0                         60 exec        106
> ttyrawch
> hdisk1     0.0     0.0    0   Client Server NFS/s    0 rcvint        0
> ttycanch
> hdisk2     2.4     2.7    0      0.0    0.0 calls    0 xmtint      211
> ttyoutch
> hdisk3     0.0     2.7    0      0.0    0.0 retry    0 mdmint
> hdisk4     0.0   47.7    0      0.0    0.0 getattr
> hdisk5     0.0   12.5    0      0.0    0.0 lookup       Netw   read
> write
> kB/s
> hdisk6     0.0     0.0    0      0.0    0.0 read        lo0     0.0
> 0.0
> hdisk7     0.0     5.4    0      0.0    0.0 write       en0     7.2
> 37.1
> hdisk8     0.0   136.2    0      0.0    0.0 other       en1     0.0
> 0.0
> hdisk9     2.4    0.0     0

--
Urban A. Haas
CEO - Urban Technology, Inc.
Minneapolis, MN  USA
Phone: (952) 595-8810    Fax: (952) 595-8710

Web: http://www.urbantechnology.com

This e-mail was composed of 100% recycled bits.

 
 
 

1. network performance problem

Hello all,

I was running a ipmasquerade network, with two machines I will call them
insideA, and insideB from inside accessing one internet connection on
machine I will call gatewayC.  For some reason everything is sooo slow
lately.  I mean from the two machines insideA/insideB accessing gatewayC
as well as machine gatewayC accessing the Internet.  Trying to access the
web from insideA/insideB via gatewayC is just extremely slow.  I tried
disabling most services on gatewayC and still things has not speed up.
I just moved to ipchains in hope of solving this problem, it didn't work.

The only other thing I am concerned of is that the machine has been
hacked into before while I left it on for a few days.  ie last log showed
someone ftp'ed into it from an unknown location.  Also log showed that
they tried to access start imapd2 repeatedly. They also attempted SYN packet
attacked at around the same time.  Hopefully he didn't leave any running
funny program that bog down the system.  :-)

I also suspect the hub that I am using maybe defective but have no extra
hub to test out this theory.

Are there any tools out there that will show me where the bottle neck might
be?  like whether it is the hub? or the linux box itself?

I appreciate any help in resolving this mess.  Please email if possible.

INFOS:  I am now running kernel 2.2.9, basic linux installation, with
        apache server. CPU: AMD 200, 16MB RAM  (before you say RAM!, I
        have to say it worked just fine before with a swap space).

2. LFS and lilo

3. Network performance Problem

4. Timer chip/card wanted

5. RS6000 network performance problem

6. VMPower G3 Upgrade Card and LINUX

7. Network performance problems under Linux, but not Windows

8. ZIP drive

9. network performance problems...

10. 3COM network performance problems with Redhat 5.0

11. Networking performance problem

12. One Way Network Performance Problem

13. PPP Networking Performance Problems