Thread View: demon.ip.support.nt
12 messages
12 total messages
Started by Duncan
Tue, 05 Mar 2002 09:40
Why does windows2000 ping have a resolution of 10ms?
Author: Duncan
Date: Tue, 05 Mar 2002 09:40
Date: Tue, 05 Mar 2002 09:40
14 lines
1611 bytes
1611 bytes
I just noticed all the pings from a windows2000 pc on a LAN are always n*10 ms i.e Pinging 158.152.1.222 with 32 bytes of data: Reply from 158.152.1.222: bytes=32 time=50ms TTL=254 Reply from 158.152.1.222: bytes=32 time=40ms TTL=254 Reply from 158.152.1.222: bytes=32 time=40ms TTL=254 Reply from 158.152.1.222: bytes=32 time=40ms TTL=254 Looking at other peoples trace routes in winnt it seems this is common but no one seems to ask why. Is windows adding some latency or is it just that its ping timing is inaccurate? -- Duncan
Re: Why does windows2000 ping have a resolution of 10ms?
Author: Bob Cousins
Date: Wed, 06 Mar 2002 09:33
Date: Wed, 06 Mar 2002 09:33
13 lines
1694 bytes
1694 bytes
Nick Webster wrote: >From here, using 2000 and XP, pinging localhost, other LAN machines, >default gateway, and hosts on the Net, all produce pings with various >times that are not n*10. Thats funny, all the ping times I have ever seen from many NT and 2000 boxes have been in 10ms granularity. I assumed this was because the resolution of the Windows Timer function is 10ms. I wonder what is different about your systems? -- Bob Cousins. Please remove $NOSPAM$ to reply.
Re: Why does windows2000 ping have a resolution of 10ms?
Author: Bob Cousins
Date: Wed, 06 Mar 2002 11:51
Date: Wed, 06 Mar 2002 11:51
14 lines
1857 bytes
1857 bytes
Nick Webster wrote: >All I can think of is it's fixed in XP, and maybe one of the service >packs for 2000? Hmm, thats odd, my other W2K system does use 1ms timer. Both are supposed to be SP2. Perhaps it depends on the hardware or kernel used. Looking on the Knowledge Base, there are quite a few bugs on ping, so it may have been fixed in some revision. Ping also had the classic 49.7 day failure. -- Bob Cousins. Please remove $NOSPAM$ to reply.
Re: Why does windows2000 ping have a resolution of 10ms?
Author: John Devereux
Date: Wed, 06 Mar 2002 14:07
Date: Wed, 06 Mar 2002 14:07
15 lines
1806 bytes
1806 bytes
On Wed, 06 Mar 2002 11:51:50 +0000, Bob Cousins <bob@$NOSPAM$lintilla.demon.co.uk> wrote: >Nick Webster wrote: > >Looking on the Knowledge Base, there are quite a few bugs on ping, so >it may have been fixed in some revision. Ping also had the classic >49.7 day failure. Pings should not take that long, even with Demon... -- John Devereux john@devereux.demon.co.uk
Re: Why does windows2000 ping have a resolution of 10ms?
Author: Nigel Kendrick
Date: Wed, 06 Mar 2002 23:39
Date: Wed, 06 Mar 2002 23:39
19 lines
1875 bytes
1875 bytes
Bob Cousins wrote: > Nick Webster wrote: > >>From here, using 2000 and XP, pinging localhost, other LAN machines, > >>default gateway, and hosts on the Net, all produce pings with various >>times that are not n*10. >> > > Thats funny, all the ping times I have ever seen from many NT and 2000 > boxes have been in 10ms granularity. I assumed this was because the > resolution of the Windows Timer function is 10ms. I wonder what is > different about your systems? > NIC drivers? NK
Re: Why does windows2000 ping have a resolution of 10ms?
Author: Nigel Kendrick
Date: Wed, 06 Mar 2002 23:40
Date: Wed, 06 Mar 2002 23:40
20 lines
2003 bytes
2003 bytes
John Devereux wrote: > On Wed, 06 Mar 2002 11:51:50 +0000, Bob Cousins > <bob@$NOSPAM$lintilla.demon.co.uk> wrote: > > >>Nick Webster wrote: >> >>Looking on the Knowledge Base, there are quite a few bugs on ping, so >>it may have been fixed in some revision. Ping also had the classic >>49.7 day failure. >> > > Pings should not take that long, even with Demon... > > "...unless they're via Demon..." shurely!? NK
Re: Why does windows2000 ping have a resolution of 10ms?
Author: Duncan
Date: Sat, 09 Mar 2002 12:10
Date: Sat, 09 Mar 2002 12:10
8 lines
1481 bytes
1481 bytes
Nigel Kendrick <""@nkendrick.?> wrote > >NIC drivers? Mines a D-link DFE-530TX and I get the 10ms granularity. Maybe someone with 1ms can tell us with what NIC? -- Duncan
Re: Why does windows2000 ping have a resolution of 10ms?
Author: Bob Cousins
Date: Sat, 09 Mar 2002 16:15
Date: Sat, 09 Mar 2002 16:15
14 lines
1718 bytes
1718 bytes
Duncan wrote: >Nigel Kendrick <""@nkendrick.?> wrote >> >>NIC drivers? > >Mines a D-link DFE-530TX and I get the 10ms granularity. Maybe someone >with 1ms can tell us with what NIC? I can't imagine why the NIC would make a difference, surely the ping application is responsible for the timing? -- Bob Cousins. Please remove $NOSPAM$ to reply.
Re: Why does windows2000 ping have a resolution of 10ms?
Author: Simon
Date: Sat, 09 Mar 2002 21:16
Date: Sat, 09 Mar 2002 21:16
42 lines
3063 bytes
3063 bytes
* Bob Cousins <bob@$nospam$lintilla.demon.co.uk> wrote:
> Duncan wrote:
>>Nigel Kendrick <""@nkendrick.?> wrote
>>>
>>>NIC drivers?
>>
>>Mines a D-link DFE-530TX and I get the 10ms granularity. Maybe someone
>>with 1ms can tell us with what NIC?
>
> I can't imagine why the NIC would make a difference, surely the ping
> application is responsible for the timing?
My results seem to suggest that the ping application isn't responsible for the
timing. My results go from ping times accurate to 1 decimal place in
netkit-base-0.10 to almost accurate to 10ms in netkit-base-0.17 (apart from
the first ping):
An Intel EtherExpress Pro 100 on Linux 2.4.5 using ping from netkit-base-0.17:
64 octets from 158.152.1.222: icmp_seq=0 ttl=255 time=155.4 ms
64 octets from 158.152.1.222: icmp_seq=1 ttl=255 time=140.0 ms
64 octets from 158.152.1.222: icmp_seq=2 ttl=255 time=150.0 ms
...results carry on accurate to 10ms until the ping is stopped
A 3com 905b on Linux 2.4.17 using ping from netkit-base-0.10:
64 bytes from 158.152.1.222: icmp_seq=0 ttl=254 time=150.9 ms
64 bytes from 158.152.1.222: icmp_seq=1 ttl=254 time=151.5 ms
64 bytes from 158.152.1.222: icmp_seq=2 ttl=254 time=151.3 ms
A 3com 3C574-TX on FreeBSD 4.3, looks to be using ping version 1.52.2.2:
64 bytes from 158.152.1.222: icmp_seq=0 ttl=254 time=178.932 ms
64 bytes from 158.152.1.222: icmp_seq=1 ttl=254 time=186.622 ms
64 bytes from 158.152.1.222: icmp_seq=2 ttl=254 time=156.533 ms
--
Simon <simon@sis-domain.demon.co.uk> **** PGP: 099977D0
"We demand rigidly defined areas of doubt and uncertainty."
- Douglas Adams
Re: Why does windows2000 ping have a resolution of 10ms?
Author: Simon
Date: Sat, 09 Mar 2002 21:50
Date: Sat, 09 Mar 2002 21:50
16 lines
2167 bytes
2167 bytes
* Simon <usenet@sis-domain.demon.co.uk> wrote:
> An Intel EtherExpress Pro 100 on Linux 2.4.5 using ping from netkit-base-0.17:
>
> 64 octets from 158.152.1.222: icmp_seq=0 ttl=255 time=155.4 ms
> 64 octets from 158.152.1.222: icmp_seq=1 ttl=255 time=140.0 ms
> 64 octets from 158.152.1.222: icmp_seq=2 ttl=255 time=150.0 ms
> ...results carry on accurate to 10ms until the ping is stopped
To follow up my own post, this is over the PPP stack (using pppd 2.4.1) and
not the NIC. Must acquire more caffeine...
--
Simon <simon@sis-domain.demon.co.uk> **** PGP: 099977D0
"We demand rigidly defined areas of doubt and uncertainty."
- Douglas Adams
Re: Why does windows2000 ping have a resolution of 10ms?
Author: Bob Cousins
Date: Sun, 10 Mar 2002 00:52
Date: Sun, 10 Mar 2002 00:52
27 lines
2568 bytes
2568 bytes
Simon wrote: >* Simon <usenet@sis-domain.demon.co.uk> wrote: >> An Intel EtherExpress Pro 100 on Linux 2.4.5 using ping from netkit-base-0.17: >> >> 64 octets from 158.152.1.222: icmp_seq=0 ttl=255 time=155.4 ms >> 64 octets from 158.152.1.222: icmp_seq=1 ttl=255 time=140.0 ms >> 64 octets from 158.152.1.222: icmp_seq=2 ttl=255 time=150.0 ms >> ...results carry on accurate to 10ms until the ping is stopped > >To follow up my own post, this is over the PPP stack (using pppd 2.4.1) and >not the NIC. Must acquire more caffeine... Ooops.. Actually, I am fairly certain the ping application calculates the time interval, but it is possible the NIC driver affects the resolution, for example depending on how interrupts get handled and how the request queues are managed. Linux source code is obviously freely available, if you look at the source to ping you will see it uses the system timer. The Unix socket interface does not carry timestamps of packets. Winsock derives from the Unix socket API, so is similar. -- Bob Cousins. Please remove $NOSPAM$ to reply.
Re: Why does windows2000 ping have a resolution of 10ms?
Author: Simon
Date: Sun, 10 Mar 2002 09:41
Date: Sun, 10 Mar 2002 09:41
40 lines
3413 bytes
3413 bytes
* Bob Cousins <bob@$nospam$lintilla.demon.co.uk> wrote:
> Simon wrote:
>>* Simon <usenet@sis-domain.demon.co.uk> wrote:
>>> An Intel EtherExpress Pro 100 on Linux 2.4.5 using ping from
>>> netkit-base-0.17:
>>>
>>> 64 octets from 158.152.1.222: icmp_seq=0 ttl=255 time=155.4 ms
>>> 64 octets from 158.152.1.222: icmp_seq=1 ttl=255 time=140.0 ms
>>> 64 octets from 158.152.1.222: icmp_seq=2 ttl=255 time=150.0 ms
>>> ...results carry on accurate to 10ms until the ping is stopped
>>
>>To follow up my own post, this is over the PPP stack (using pppd 2.4.1) and
>>not the NIC. Must acquire more caffeine...
>
> Ooops..
>
> Actually, I am fairly certain the ping application calculates the time
> interval, but it is possible the NIC driver affects the resolution,
> for example depending on how interrupts get handled and how the
> request queues are managed.
Looking at the ping source code, it looks as though the application uses the
system timer to calculate the reply times. From this, I think the most
likely causes might be either interrupts or drivers.
> Linux source code is obviously freely available, if you look at the
> source to ping you will see it uses the system timer. The Unix socket
> interface does not carry timestamps of packets. Winsock derives from
> the Unix socket API, so is similar.
This agrees with my knowledge of the TCP & IP protocols. If I can get my head
around the PPP and/or serial port drivers in the kernel, I'll see if those
would affect the timing in any way (the Linux box with more accurate timing
is running a newer version of the kernel).
--
Simon <simon@sis-domain.demon.co.uk> **** PGP: 099977D0
"We demand rigidly defined areas of doubt and uncertainty."
- Douglas Adams
Thread Navigation
This is a paginated view of messages in the thread with full content displayed inline.
Messages are displayed in chronological order, with the original post highlighted in green.
Use pagination controls to navigate through all messages in large threads.
Back to All Threads