(Ypu may have seen this in comp.unix.solaris, I am casting my net wider
still...)

I have a program (run from inetd, but that shouldn't matter) that does
'popen("ssh host /bin/sh", "w")' and feeds a simple script down the line.

After about an hour, the ypbind daemon on the RH boxes dies.

The NIS server is a Solaris 8 machine.

All the non-RH (Solaris and HP-UX) systems in the NIS domain run fine.

Customer says the the RH boxes are "Redhat Enterprise Update 4", I
presume that they mean "Red Hat Enterprise Linux, version 4" and are
therefore running SELinux.

I don't know much more than that, but I will be asking questions.

My google-fu isn't shining any light anywhere, and I don't have an NIS
domain to play with.  (And I'm on holidays at the moment, so I'm living
at about 46.6Kbps...)

Suggestions, pointers, all gratefully accepted.  (Apart from "don't use
RH/NIS," I don't have that option)  (But "update your NIS server because
of x, y, and z" might be okay)

Cheers,
Gary	B-)

--
 ________________________________________
____________________________________
__
Armful of chairs: Something some people would not know
whether you were up them with or not
- Barry Humphries

[ Post a follow-up to this message ]

"Gary R. Schmidt" <grschmidt@acm.org> wrote in message
news:42jimjF1jm63dU1@individual.net

> After about an hour, the ypbind daemon on the RH boxes dies.
>
> The NIS server is a Solaris 8 machine.

What indications are in the log files, both client and server?

[ Post a follow-up to this message ]

In comp.os.linux.networking Gary R. Schmidt <grschmidt@acm.org>:
> (Ypu may have seen this in comp.unix.solaris, I am casting my net wider
> still...)

> I have a program (run from inetd, but that shouldn't matter) that does
> 'popen("ssh host /bin/sh", "w")' and feeds a simple script down the line.

> After about an hour, the ypbind daemon on the RH boxes dies.

> The NIS server is a Solaris 8 machine.

> All the non-RH (Solaris and HP-UX) systems in the NIS domain run fine.

> Customer says the the RH boxes are "Redhat Enterprise Update 4", I
> presume that they mean "Red Hat Enterprise Linux, version 4" and are
> therefore running SELinux.

No, there's no update 4 for RHEL 4.0. Sounds like update 4 for
RHEL 3.0, which is outdated. Iirc update 6 is recent, I'd update
the box to the latest (up2date) and see if this improves things
before looking deeper into the problem. You get some support with
RHEL, so you could open up a ticket to RH.

[..]

--
Michael Heiming (X-PGP-Sig > GPG-Key ID: EDD27B94)
mail: echo zvpunry@urvzvat.qr | PERL -pe 'y/a-z/n-za-m/'
#bofh excuse 340: We'll fix that in the next (upgrade, update,
patch release, service pack).

[ Post a follow-up to this message ]

ynotssor wrote:

> "Gary R. Schmidt" <grschmidt@acm.org> wrote in message
> news:42jimjF1jm63dU1@individual.net
>
> 
>
>
> What indications are in the log files, both client and server?
Waiting on the customer to supply these.

Cheers,
Gary	B-)

--
 ________________________________________
____________________________________
__
Armful of chairs: Something some people would not know
whether you were up them with or not
- Barry Humphries

[ Post a follow-up to this message ]

Michael Heiming wrote:

> In comp.os.linux.networking Gary R. Schmidt <grschmidt@acm.org>:
>
[SNIP]
>
>
> No, there's no update 4 for RHEL 4.0. Sounds like update 4 for
> RHEL 3.0, which is outdated. Iirc update 6 is recent, I'd update
> the box to the latest (up2date) and see if this improves things
> before looking deeper into the problem. You get some support with
> RHEL, so you could open up a ticket to RH.
>
Thanks for that.  I'll pass it on and see if the customer screams or
not.  (They appear somewhat naive).

Cheers,
Gary	B-)

--
 ________________________________________
____________________________________
__
Armful of chairs: Something some people would not know
whether you were up them with or not
- Barry Humphries

[ Post a follow-up to this message ]

Gary R. Schmidt wrote:

> Michael Heiming wrote:
> 
> [SNIP]
> 
> Thanks for that.  I'll pass it on and see if the customer screams or
> not.  (They appear somewhat naive).
>
Sigh.

Customer has decided to go back to a previous version, when I get back
from hols I'll presumably have to set up an NIS domain that simulates
theirs and do some testing.

No doubt I shall return!

Cheers,
Gary	B-)

--
 ________________________________________
____________________________________
__
Armful of chairs: Something some people would not know
whether you were up them with or not
- Barry Humphries

[ Post a follow-up to this message ]