This is the mail archive of the cygwin@cygwin.com mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Executing commands is VERY slow when logged into an NT domain....


On Thu, Jun 27, 2002 at 07:26:49PM -0700, Rick Rankin wrote:
>
>--- Christopher Faylor <cgf@redhat.com> wrote:
>> On Wed, Jun 26, 2002 at 10:36:19PM -0700, Rick Rankin wrote:
>> >
>> >--- Christopher Faylor <cgf@redhat.com> wrote:
>> >> Out of curiousity, does mounting your /bin and /usr/bin directories with
>> the
>> >> -X
>> >> option "fix" anything?
>> >> 
>> >> mount -f -b -X c:/cygwin/bin /bin
>> >> mount -f -b -X c:/cygwin/bin /usr/bin
>> >> 
>> >> where c:\cygwin is the location of your root, of course.
>> >
>> >It turns out that the -X option does help quite a bit. Here are some rough
>> >results based on using a stopwatch to time the interval from when I
>> >double-click the bash icon on my desktop until I see a prompt:
>> >
>> >20020626 snapshot,   no -X: 25.2 sec
>> >20020626 snapshot, with -X:  6.7 sec
>> >1.3.10-1,            no -X:  2.1 sec
>> >
>> >I didn't retry the 0625 snapshot, but it ran about the same speed without
>> the 
>> >-X option.
>> >
>> >Does that narrow things down any?
>> 
>> Some, but it is still mystifying.  I would have expected -X to actually
>> completely
>> mask the problem.  There are some more changes in CVS.  I'll regen a snapshot
>> as
>> soon as Corinna checks in her latest change.
>> 
>
>Here's some more timeing results with the 20020627 snapshot that I just
>downloaded at about 7:00 PM MST (in case there was more that one incarnation of
>the 0627 snapshot). There's not much difference:
>
>20020627 snapshot,   no -X: 24.8 sec
>20020627 snapshot, with -X:  6.7 sec
>
>FWIW, I have a small app called NetSpeed that sits in the system tray and
>measures network throughput. The icon in the tray is active and gives a good
>indication of network activity levels. When I start bash (or any command from
>bash), NetSpeed indicates that there is definitely an amount of network
>activity occurring with the 1.3.11 dlls that doesn't occur with 1.3.10. This
>would seem to confirm the theory that the 1.3.11 dll is trying to contact the
>domain controller. Also, task manager shows that there is no significant CPU
>activity during this time, which there wouldn't be if it's blocked waiting for
>a response.

Could you run strace on a short /bin/sh (not bash) session:

  strace -ostrace.out sh
  /bin/pwd
  exit

and send it here?

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]