samba - Oct 2001 - smb_trans2_request: result=-104, setting invalid

Why do I keep getting these messages?
(from /var/log/messages)
I use RedHat 7.1 and have a coupple of shares on the machine on a bizzy
samba network.

Oct 10 04:03:50 n243 kernel: smb_trans2_request: result=-104, setting
invalid
Oct 10 04:03:50 n243 kernel: smb_retry: successful, new pid=5506,
generation=47
Oct 10 04:44:03 n243 kernel: smb_trans2_request: result=-104, setting
invalid
Oct 10 04:44:03 n243 kernel: smb_retry: successful, new pid=5506,
generation=48
Oct 10 09:03:22 n243 kernel: smb_trans2_request: result=-104, setting
invalid
Oct 10 09:03:22 n243 kernel: smb_retry: successful, new pid=5506,
generation=49
Oct 10 09:19:33 n243 kernel: smb_trans2_request: result=-104, setting
invalid
Oct 10 09:19:34 n243 kernel: smb_retry: successful, new pid=5506,
generation=50
Oct 10 09:52:26 n243 kernel: smb_trans2_request: result=-104, setting
invalid
Oct 10 09:52:26 n243 kernel: smb_retry: successful, new pid=5506,
generation=51
Oct 10 11:01:40 n243 kernel: smb_trans2_request: result=-104, setting
invalid
Oct 10 11:01:40 n243 kernel: smb_retry: successful, new pid=5506,
generation=52

Could these above messages have something to do whit this (also shows a
coupple of times in the log)

Oct  9 23:13:17 n243 kernel:   Flags; bus-master 1, dirty 7289149(13)
current 7289149(13)
Oct  9 23:13:17 n243 kernel:   Transmit list 00000000 vs. c11fb540.

On Thu, 11 Oct 2001, Martin Eriksson wrote:
> Why do I keep getting these messages?
> (from /var/log/messages)
> I use RedHat 7.1 and have a coupple of shares on the machine on a bizzy
> samba network.
> 
> Oct 10 04:03:50 n243 kernel: smb_trans2_request: result=-104, setting
> invalid
> Oct 10 04:03:50 n243 kernel: smb_retry: successful, new pid=5506,
> generation=47
> Oct 10 04:44:03 n243 kernel: smb_trans2_request: result=-104, setting
> invalid
> Oct 10 04:44:03 n243 kernel: smb_retry: successful, new pid=5506,
> generation=48
-104 == -ECONNRESET, or "Connection reset by peer"

The server has disconnected because the mount was idle for too long
(default is something like 10-15 minutes). However, smb_retry is then
called on the next access to the mount and it successfully mounts the
share again.
> Oct 10 11:01:40 n243 kernel: smb_trans2_request: result=-104, setting
> invalid
> Oct 10 11:01:40 n243 kernel: smb_retry: successful, new pid=5506,
> generation=52
And it keeps doing that (generation is the number of successful remounts).
None of these are errors.

With older kernels and older samba versions this would cause smbmount to
crash and that in turn creates EIO error respones from all accesses to the
mount.


Some people run scripts to avoid disconnects, try looking in the samba
mail archive at marc.theaimsgroup.com. You can also change the registry on 
the server and increase the 'autodisconnect' parameter.

/Urban

I'm having problems with smb connections either timing out or becoming
corrupted after about 5 minutes of inactivity???  We mount all our user
homedirs (smb shares) through pam_mount at login.  The workaround is to
logout and login again.

/var/log/messages has a line that looks related: kernel:
smb_trans2_request: result=-104, setting invalid

Please help!