Two servers breaking connections (Long, challenging)

Here is the setup. We have 4 Windows 2000 servers at this site with about
30 XP workstations. This is a site within a 3 site domain. There is one DC
at each site. But I'm only having a problem with this site so I don't think
the sites are the problem. One of our Win2000 servers has gobs of hard
drive space so I set up the other 3 to backup to it, that is, they are just
backing up to files - not a tape drive. Each backup runs 2 hours apart and
only takes about a half hour so they are not overlapping. But this still
isn't the problem as this runs smoothly. The problem is 1 server is now
loosing it's connection to this server every couple of minutes and after a
few minutes it connects back up. Neither of these are the DC, they are just
plain servers. Nothing has changed as far as I can tell. On the backup
server there are no errors in the logs but on the server that's sending the
file I get numerous Event ID 3034 with the last part of the data being
c0000133 which, according to numerous sources, indicates a time error of
some kind. I've checked the clocks and they are spot on to the DC and I set
up a policy to allow a 10 minute deviation of clocks, but like I said, they
are spot on. I've even checked the CMOS times and dates and time zones and
cannot find a single second off between the two.

If I go into the network neighborhood on the sending server and click on the
backup server when this is happening it asks for a password and I can login
with the domain admin account and nothing else. Once I do this the
connection is back up. I've checked all DNS and WINS records, can't find a
name or IP conflict, cleared caches, and the 3034 error is the only error
(warning, really) in the log of the sending server, the backup server has
numerous (seconds apart) event ID 100 from MSFTPSVS which says "The server
was unable to logon the Windows NT account "Administrator" due to the
follwoing error: Logon Failure: unknown user name or bad password. The data
is the error code." with the last entry in the data being 0000052e. Is this
the sending server failing to log on? Or a service failing to start on the
backup server? The other two servers that send backups are not having a
problem at all. Sometimes the connection will last a day or two but mostly
the error pops up every 20-30 minutes, which doesn't correspond to anything
that I can find. No other servers anywhere are having any problem, it's
just between these two. IT'S NOT THE CLOCK for those of you about to
suggest to reset the time. I've done this and checked this many, many
times. The only transfer between these two servers is the backup which is a
half-hour at the most, once a day.

This really is a smooth-running network and I've been chasing this error for
a few months now. Any suggestions at all? Anything else I can monitor?
Neither of these accesses the internet and yes, I've checked for viruses and
spyware.

Thanks to anyone who got this far!

Maybe you could try backing up in the opposite direction. Instead of
backing up files from Server A onto server B, try backing up files from
server B onto server A. See if the connection drops out then. I assume
you tried replacing the network cards involved. You could even try a
direct crossover connection cable between them to rule out a hub or
switch problem.


RB wrote:
> Here is the setup. We have 4 Windows 2000 servers at this site with about
> 30 XP workstations. This is a site within a 3 site domain. There is one DC
> at each site. But I'm only having a problem with this site so I don't think
> the sites are the problem. One of our Win2000 servers has gobs of hard
> drive space so I set up the other 3 to backup to it, that is, they are just
> backing up to files - not a tape drive. Each backup runs 2 hours apart and
> only takes about a half hour so they are not overlapping. But this still
> isn't the problem as this runs smoothly. The problem is 1 server is now
> loosing it's connection to this server every couple of minutes and after a
> few minutes it connects back up. Neither of these are the DC, they are just
> plain servers. Nothing has changed as far as I can tell. On the backup
> server there are no errors in the logs but on the server that's sending the
> file I get numerous Event ID 3034 with the last part of the data being
> c0000133 which, according to numerous sources, indicates a time error of
> some kind. I've checked the clocks and they are spot on to the DC and I set
> up a policy to allow a 10 minute deviation of clocks, but like I said, they
> are spot on. I've even checked the CMOS times and dates and time zones and
> cannot find a single second off between the two.
>
> If I go into the network neighborhood on the sending server and click on the
> backup server when this is happening it asks for a password and I can login
> with the domain admin account and nothing else. Once I do this the
> connection is back up. I've checked all DNS and WINS records, can't find a
> name or IP conflict, cleared caches, and the 3034 error is the only error
> (warning, really) in the log of the sending server, the backup server has
> numerous (seconds apart) event ID 100 from MSFTPSVS which says "The server
> was unable to logon the Windows NT account "Administrator" due to the
> follwoing error: Logon Failure: unknown user name or bad password. The data
> is the error code." with the last entry in the data being 0000052e. Is this
> the sending server failing to log on? Or a service failing to start on the
> backup server? The other two servers that send backups are not having a
> problem at all. Sometimes the connection will last a day or two but mostly
> the error pops up every 20-30 minutes, which doesn't correspond to anything
> that I can find. No other servers anywhere are having any problem, it's
> just between these two. IT'S NOT THE CLOCK for those of you about to
> suggest to reset the time. I've done this and checked this many, many
> times. The only transfer between these two servers is the backup which is a
> half-hour at the most, once a day.
>
> This really is a smooth-running network and I've been chasing this error for
> a few months now. Any suggestions at all? Anything else I can monitor?
> Neither of these accesses the internet and yes, I've checked for viruses and
> spyware.
>
> Thanks to anyone who got this far!

I did try 1 network card (the backup server is built-in and no other server
is having this problem) and connected the servers on another switch in the
stack. I don't have two extra cards so at the moment can't try to connect
them directly, and I can't use the current connections because these servers
are critical to the network. I also don't think it's a Backup Exec issue as
1) it only runs 1/2 hour a day and 2) the connection is dropped all
throughout the day.

I am trying to tweak log on permissions in both servers and on the DC in the
policies that would apply. It may be a log-in locally right
misconfiguration but I've just done that so only time will tell now.

Thanks for the suggestions. I may try a reverse backup just to see what
happens.


"Greg O" <(E-Mail Removed)> wrote in message
news:(E-Mail Removed) oups.com...
> Maybe you could try backing up in the opposite direction. Instead of
> backing up files from Server A onto server B, try backing up files from
> server B onto server A. See if the connection drops out then. I assume
> you tried replacing the network cards involved. You could even try a
> direct crossover connection cable between them to rule out a hub or
> switch problem.
>
>
> RB wrote:
>> Here is the setup. We have 4 Windows 2000 servers at this site with
>> about
>> 30 XP workstations. This is a site within a 3 site domain. There is one
>> DC
>> at each site. But I'm only having a problem with this site so I don't
>> think
>> the sites are the problem. One of our Win2000 servers has gobs of hard
>> drive space so I set up the other 3 to backup to it, that is, they are
>> just
>> backing up to files - not a tape drive. Each backup runs 2 hours apart
>> and
>> only takes about a half hour so they are not overlapping. But this still
>> isn't the problem as this runs smoothly. The problem is 1 server is now
>> loosing it's connection to this server every couple of minutes and after
>> a
>> few minutes it connects back up. Neither of these are the DC, they are
>> just
>> plain servers. Nothing has changed as far as I can tell. On the backup
>> server there are no errors in the logs but on the server that's sending
>> the
>> file I get numerous Event ID 3034 with the last part of the data being
>> c0000133 which, according to numerous sources, indicates a time error of
>> some kind. I've checked the clocks and they are spot on to the DC and I
>> set
>> up a policy to allow a 10 minute deviation of clocks, but like I said,
>> they
>> are spot on. I've even checked the CMOS times and dates and time zones
>> and
>> cannot find a single second off between the two.
>>
>> If I go into the network neighborhood on the sending server and click on
>> the
>> backup server when this is happening it asks for a password and I can
>> login
>> with the domain admin account and nothing else. Once I do this the
>> connection is back up. I've checked all DNS and WINS records, can't find
>> a
>> name or IP conflict, cleared caches, and the 3034 error is the only error
>> (warning, really) in the log of the sending server, the backup server has
>> numerous (seconds apart) event ID 100 from MSFTPSVS which says "The
>> server
>> was unable to logon the Windows NT account "Administrator" due to the
>> follwoing error: Logon Failure: unknown user name or bad password. The
>> data
>> is the error code." with the last entry in the data being 0000052e. Is
>> this
>> the sending server failing to log on? Or a service failing to start on
>> the
>> backup server? The other two servers that send backups are not having a
>> problem at all. Sometimes the connection will last a day or two but
>> mostly
>> the error pops up every 20-30 minutes, which doesn't correspond to
>> anything
>> that I can find. No other servers anywhere are having any problem, it's
>> just between these two. IT'S NOT THE CLOCK for those of you about to
>> suggest to reset the time. I've done this and checked this many, many
>> times. The only transfer between these two servers is the backup which
>> is a
>> half-hour at the most, once a day.
>>
>> This really is a smooth-running network and I've been chasing this error
>> for
>> a few months now. Any suggestions at all? Anything else I can monitor?
>> Neither of these accesses the internet and yes, I've checked for viruses
>> and
>> spyware.
>>
>> Thanks to anyone who got this far!
>

Assuming the servers are on the same subnet there may be a workaround.
You could route the server traffic through one of the other servers
using RRAS. So if there are problems with connecting servers A and B
you make a static route from A through server C. To do this you shold
be able to make a static route on Server A's network card, telling it
the route to B is using Server C as a gateway. Since C works ok with A
and B (if I understand you) then this should not drop out. I know it's
not a real solution but the extra traffic through C may not overload
it.

I suppose you could always back up server A on C, and then include that
backup when C backs up on B. Complicated but less time consuming than
reinstalling A.

RB wrote:
> I did try 1 network card (the backup server is built-in and no other server
> is having this problem) and connected the servers on another switch in the
> stack. I don't have two extra cards so at the moment can't try to connect
> them directly, and I can't use the current connections because these servers
> are critical to the network. I also don't think it's a Backup Exec issue as
> 1) it only runs 1/2 hour a day and 2) the connection is dropped all
> throughout the day.
>
> I am trying to tweak log on permissions in both servers and on the DC in the
> policies that would apply. It may be a log-in locally right
> misconfiguration but I've just done that so only time will tell now.
>
> Thanks for the suggestions. I may try a reverse backup just to see what
> happens.
>
>
> "Greg O" <(E-Mail Removed)> wrote in message
> news:(E-Mail Removed) oups.com...
> > Maybe you could try backing up in the opposite direction. Instead of
> > backing up files from Server A onto server B, try backing up files from
> > server B onto server A. See if the connection drops out then. I assume
> > you tried replacing the network cards involved. You could even try a
> > direct crossover connection cable between them to rule out a hub or
> > switch problem.
> >
> >
> > RB wrote:
> >> Here is the setup. We have 4 Windows 2000 servers at this site with
> >> about
> >> 30 XP workstations. This is a site within a 3 site domain. There is one
> >> DC
> >> at each site. But I'm only having a problem with this site so I don't
> >> think
> >> the sites are the problem. One of our Win2000 servers has gobs of hard
> >> drive space so I set up the other 3 to backup to it, that is, they are
> >> just
> >> backing up to files - not a tape drive. Each backup runs 2 hours apart
> >> and
> >> only takes about a half hour so they are not overlapping. But this still
> >> isn't the problem as this runs smoothly. The problem is 1 server is now
> >> loosing it's connection to this server every couple of minutes and after
> >> a
> >> few minutes it connects back up. Neither of these are the DC, they are
> >> just
> >> plain servers. Nothing has changed as far as I can tell. On the backup
> >> server there are no errors in the logs but on the server that's sending
> >> the
> >> file I get numerous Event ID 3034 with the last part of the data being
> >> c0000133 which, according to numerous sources, indicates a time error of
> >> some kind. I've checked the clocks and they are spot on to the DC and I
> >> set
> >> up a policy to allow a 10 minute deviation of clocks, but like I said,
> >> they
> >> are spot on. I've even checked the CMOS times and dates and time zones
> >> and
> >> cannot find a single second off between the two.
> >>
> >> If I go into the network neighborhood on the sending server and click on
> >> the
> >> backup server when this is happening it asks for a password and I can
> >> login
> >> with the domain admin account and nothing else. Once I do this the
> >> connection is back up. I've checked all DNS and WINS records, can't find
> >> a
> >> name or IP conflict, cleared caches, and the 3034 error is the only error
> >> (warning, really) in the log of the sending server, the backup server has
> >> numerous (seconds apart) event ID 100 from MSFTPSVS which says "The
> >> server
> >> was unable to logon the Windows NT account "Administrator" due to the
> >> follwoing error: Logon Failure: unknown user name or bad password. The
> >> data
> >> is the error code." with the last entry in the data being 0000052e. Is
> >> this
> >> the sending server failing to log on? Or a service failing to start on
> >> the
> >> backup server? The other two servers that send backups are not having a
> >> problem at all. Sometimes the connection will last a day or two but
> >> mostly
> >> the error pops up every 20-30 minutes, which doesn't correspond to
> >> anything
> >> that I can find. No other servers anywhere are having any problem, it's
> >> just between these two. IT'S NOT THE CLOCK for those of you about to
> >> suggest to reset the time. I've done this and checked this many, many
> >> times. The only transfer between these two servers is the backup which
> >> is a
> >> half-hour at the most, once a day.
> >>
> >> This really is a smooth-running network and I've been chasing this error
> >> for
> >> a few months now. Any suggestions at all? Anything else I can monitor?
> >> Neither of these accesses the internet and yes, I've checked for viruses
> >> and
> >> spyware.
> >>
> >> Thanks to anyone who got this far!
> >

Perhaps you could also not use Kerberos, as a time difference is only
significant for Kerberos. Unless you are worried about someone packet
sniffing on the network Kerberos might not be needed anyway.

(E-Mail Removed) wrote:
> Assuming the servers are on the same subnet there may be a workaround.
> You could route the server traffic through one of the other servers
> using RRAS. So if there are problems with connecting servers A and B
> you make a static route from A through server C. To do this you shold
> be able to make a static route on Server A's network card, telling it
> the route to B is using Server C as a gateway. Since C works ok with A
> and B (if I understand you) then this should not drop out. I know it's
> not a real solution but the extra traffic through C may not overload
> it.
>
> I suppose you could always back up server A on C, and then include that
> backup when C backs up on B. Complicated but less time consuming than
> reinstalling A.
>
> RB wrote:
> > I did try 1 network card (the backup server is built-in and no other server
> > is having this problem) and connected the servers on another switch in the
> > stack. I don't have two extra cards so at the moment can't try to connect
> > them directly, and I can't use the current connections because these servers
> > are critical to the network. I also don't think it's a Backup Exec issue as
> > 1) it only runs 1/2 hour a day and 2) the connection is dropped all
> > throughout the day.
> >
> > I am trying to tweak log on permissions in both servers and on the DC in the
> > policies that would apply. It may be a log-in locally right
> > misconfiguration but I've just done that so only time will tell now.
> >
> > Thanks for the suggestions. I may try a reverse backup just to see what
> > happens.
> >
> >
> > "Greg O" <(E-Mail Removed)> wrote in message
> > news:(E-Mail Removed) oups.com...
> > > Maybe you could try backing up in the opposite direction. Instead of
> > > backing up files from Server A onto server B, try backing up files from
> > > server B onto server A. See if the connection drops out then. I assume
> > > you tried replacing the network cards involved. You could even try a
> > > direct crossover connection cable between them to rule out a hub or
> > > switch problem.
> > >
> > >
> > > RB wrote:
> > >> Here is the setup. We have 4 Windows 2000 servers at this site with
> > >> about
> > >> 30 XP workstations. This is a site within a 3 site domain. There is one
> > >> DC
> > >> at each site. But I'm only having a problem with this site so I don't
> > >> think
> > >> the sites are the problem. One of our Win2000 servers has gobs of hard
> > >> drive space so I set up the other 3 to backup to it, that is, they are
> > >> just
> > >> backing up to files - not a tape drive. Each backup runs 2 hours apart
> > >> and
> > >> only takes about a half hour so they are not overlapping. But this still
> > >> isn't the problem as this runs smoothly. The problem is 1 server is now
> > >> loosing it's connection to this server every couple of minutes and after
> > >> a
> > >> few minutes it connects back up. Neither of these are the DC, they are
> > >> just
> > >> plain servers. Nothing has changed as far as I can tell. On the backup
> > >> server there are no errors in the logs but on the server that's sending
> > >> the
> > >> file I get numerous Event ID 3034 with the last part of the data being
> > >> c0000133 which, according to numerous sources, indicates a time error of
> > >> some kind. I've checked the clocks and they are spot on to the DC and I
> > >> set
> > >> up a policy to allow a 10 minute deviation of clocks, but like I said,
> > >> they
> > >> are spot on. I've even checked the CMOS times and dates and time zones
> > >> and
> > >> cannot find a single second off between the two.
> > >>
> > >> If I go into the network neighborhood on the sending server and click on
> > >> the
> > >> backup server when this is happening it asks for a password and I can
> > >> login
> > >> with the domain admin account and nothing else. Once I do this the
> > >> connection is back up. I've checked all DNS and WINS records, can't find
> > >> a
> > >> name or IP conflict, cleared caches, and the 3034 error is the only error
> > >> (warning, really) in the log of the sending server, the backup server has
> > >> numerous (seconds apart) event ID 100 from MSFTPSVS which says "The
> > >> server
> > >> was unable to logon the Windows NT account "Administrator" due to the
> > >> follwoing error: Logon Failure: unknown user name or bad password. The
> > >> data
> > >> is the error code." with the last entry in the data being 0000052e. Is
> > >> this
> > >> the sending server failing to log on? Or a service failing to start on
> > >> the
> > >> backup server? The other two servers that send backups are not having a
> > >> problem at all. Sometimes the connection will last a day or two but
> > >> mostly
> > >> the error pops up every 20-30 minutes, which doesn't correspond to
> > >> anything
> > >> that I can find. No other servers anywhere are having any problem, it's
> > >> just between these two. IT'S NOT THE CLOCK for those of you about to
> > >> suggest to reset the time. I've done this and checked this many, many
> > >> times. The only transfer between these two servers is the backup which
> > >> is a
> > >> half-hour at the most, once a day.
> > >>
> > >> This really is a smooth-running network and I've been chasing this error
> > >> for
> > >> a few months now. Any suggestions at all? Anything else I can monitor?
> > >> Neither of these accesses the internet and yes, I've checked for viruses
> > >> and
> > >> spyware.
> > >>
> > >> Thanks to anyone who got this far!
> > >

Already not using Kerberos. Can't backup to another server as only the
backup server has enough space on a hard drive. This may be a moot point as
since I gave logon locally rights to everyone it's been working smoothly.
The 3034 error is still popping up every couple of minutes but since all the
backups and file transfers are working smoothly it may be a totally
unrelated error. Is there any risk to leaving the logon locally right to
everyone? I may just need to back up now and find the specific right that
was needed.

Is there a monitor or something I can set up to figure out the 3034 error?
Again, the time is synched and spot on with all other servers. I've been
backwards and fowards through time settings and I can't see time being the
issue.

Thanks again for the suggestions.


<(E-Mail Removed)> wrote in message
news:(E-Mail Removed) oups.com...
>
> Perhaps you could also not use Kerberos, as a time difference is only
> significant for Kerberos. Unless you are worried about someone packet
> sniffing on the network Kerberos might not be needed anyway.
>
> (E-Mail Removed) wrote:
>> Assuming the servers are on the same subnet there may be a workaround.
>> You could route the server traffic through one of the other servers
>> using RRAS. So if there are problems with connecting servers A and B
>> you make a static route from A through server C. To do this you shold
>> be able to make a static route on Server A's network card, telling it
>> the route to B is using Server C as a gateway. Since C works ok with A
>> and B (if I understand you) then this should not drop out. I know it's
>> not a real solution but the extra traffic through C may not overload
>> it.
>>
>> I suppose you could always back up server A on C, and then include that
>> backup when C backs up on B. Complicated but less time consuming than
>> reinstalling A.
>>
>> RB wrote:
>> > I did try 1 network card (the backup server is built-in and no other
>> > server
>> > is having this problem) and connected the servers on another switch in
>> > the
>> > stack. I don't have two extra cards so at the moment can't try to
>> > connect
>> > them directly, and I can't use the current connections because these
>> > servers
>> > are critical to the network. I also don't think it's a Backup Exec
>> > issue as
>> > 1) it only runs 1/2 hour a day and 2) the connection is dropped all
>> > throughout the day.
>> >
>> > I am trying to tweak log on permissions in both servers and on the DC
>> > in the
>> > policies that would apply. It may be a log-in locally right
>> > misconfiguration but I've just done that so only time will tell now.
>> >
>> > Thanks for the suggestions. I may try a reverse backup just to see
>> > what
>> > happens.
>> >
>> >
>> > "Greg O" <(E-Mail Removed)> wrote in message
>> > news:(E-Mail Removed) oups.com...
>> > > Maybe you could try backing up in the opposite direction. Instead of
>> > > backing up files from Server A onto server B, try backing up files
>> > > from
>> > > server B onto server A. See if the connection drops out then. I
>> > > assume
>> > > you tried replacing the network cards involved. You could even try a
>> > > direct crossover connection cable between them to rule out a hub or
>> > > switch problem.
>> > >
>> > >
>> > > RB wrote:
>> > >> Here is the setup. We have 4 Windows 2000 servers at this site with
>> > >> about
>> > >> 30 XP workstations. This is a site within a 3 site domain. There
>> > >> is one
>> > >> DC
>> > >> at each site. But I'm only having a problem with this site so I
>> > >> don't
>> > >> think
>> > >> the sites are the problem. One of our Win2000 servers has gobs of
>> > >> hard
>> > >> drive space so I set up the other 3 to backup to it, that is, they
>> > >> are
>> > >> just
>> > >> backing up to files - not a tape drive. Each backup runs 2 hours
>> > >> apart
>> > >> and
>> > >> only takes about a half hour so they are not overlapping. But this
>> > >> still
>> > >> isn't the problem as this runs smoothly. The problem is 1 server is
>> > >> now
>> > >> loosing it's connection to this server every couple of minutes and
>> > >> after
>> > >> a
>> > >> few minutes it connects back up. Neither of these are the DC, they
>> > >> are
>> > >> just
>> > >> plain servers. Nothing has changed as far as I can tell. On the
>> > >> backup
>> > >> server there are no errors in the logs but on the server that's
>> > >> sending
>> > >> the
>> > >> file I get numerous Event ID 3034 with the last part of the data
>> > >> being
>> > >> c0000133 which, according to numerous sources, indicates a time
>> > >> error of
>> > >> some kind. I've checked the clocks and they are spot on to the DC
>> > >> and I
>> > >> set
>> > >> up a policy to allow a 10 minute deviation of clocks, but like I
>> > >> said,
>> > >> they
>> > >> are spot on. I've even checked the CMOS times and dates and time
>> > >> zones
>> > >> and
>> > >> cannot find a single second off between the two.
>> > >>
>> > >> If I go into the network neighborhood on the sending server and
>> > >> click on
>> > >> the
>> > >> backup server when this is happening it asks for a password and I
>> > >> can
>> > >> login
>> > >> with the domain admin account and nothing else. Once I do this the
>> > >> connection is back up. I've checked all DNS and WINS records, can't
>> > >> find
>> > >> a
>> > >> name or IP conflict, cleared caches, and the 3034 error is the only
>> > >> error
>> > >> (warning, really) in the log of the sending server, the backup
>> > >> server has
>> > >> numerous (seconds apart) event ID 100 from MSFTPSVS which says "The
>> > >> server
>> > >> was unable to logon the Windows NT account "Administrator" due to
>> > >> the
>> > >> follwoing error: Logon Failure: unknown user name or bad password.
>> > >> The
>> > >> data
>> > >> is the error code." with the last entry in the data being 0000052e.
>> > >> Is
>> > >> this
>> > >> the sending server failing to log on? Or a service failing to start
>> > >> on
>> > >> the
>> > >> backup server? The other two servers that send backups are not
>> > >> having a
>> > >> problem at all. Sometimes the connection will last a day or two but
>> > >> mostly
>> > >> the error pops up every 20-30 minutes, which doesn't correspond to
>> > >> anything
>> > >> that I can find. No other servers anywhere are having any problem,
>> > >> it's
>> > >> just between these two. IT'S NOT THE CLOCK for those of you about
>> > >> to
>> > >> suggest to reset the time. I've done this and checked this many,
>> > >> many
>> > >> times. The only transfer between these two servers is the backup
>> > >> which
>> > >> is a
>> > >> half-hour at the most, once a day.
>> > >>
>> > >> This really is a smooth-running network and I've been chasing this
>> > >> error
>> > >> for
>> > >> a few months now. Any suggestions at all? Anything else I can
>> > >> monitor?
>> > >> Neither of these accesses the internet and yes, I've checked for
>> > >> viruses
>> > >> and
>> > >> spyware.
>> > >>
>> > >> Thanks to anyone who got this far!
>> > >
>

If the 3034 error doesn't go away, try other NICs and cables before going
further afield.

If goes away after giving Everyone logon locally rights, see if you still
get good results removing Everyone and restricting it to Authenticated
Users.

HTH,
Randy Benson

"RB" <(E-Mail Removed)> wrote in message
news:44d0b119$(E-Mail Removed)...
> Already not using Kerberos. Can't backup to another server as only the
> backup server has enough space on a hard drive. This may be a moot point
> as since I gave logon locally rights to everyone it's been working
> smoothly. The 3034 error is still popping up every couple of minutes but
> since all the backups and file transfers are working smoothly it may be a
> totally unrelated error. Is there any risk to leaving the logon locally
> right to everyone? I may just need to back up now and find the specific
> right that was needed.
>
> Is there a monitor or something I can set up to figure out the 3034 error?
> Again, the time is synched and spot on with all other servers. I've been
> backwards and fowards through time settings and I can't see time being the
> issue.
>
> Thanks again for the suggestions.
>
>
> <(E-Mail Removed)> wrote in message
> news:(E-Mail Removed) oups.com...
>>
>> Perhaps you could also not use Kerberos, as a time difference is only
>> significant for Kerberos. Unless you are worried about someone packet
>> sniffing on the network Kerberos might not be needed anyway.
>>
>> (E-Mail Removed) wrote:
>>> Assuming the servers are on the same subnet there may be a workaround.
>>> You could route the server traffic through one of the other servers
>>> using RRAS. So if there are problems with connecting servers A and B
>>> you make a static route from A through server C. To do this you shold
>>> be able to make a static route on Server A's network card, telling it
>>> the route to B is using Server C as a gateway. Since C works ok with A
>>> and B (if I understand you) then this should not drop out. I know it's
>>> not a real solution but the extra traffic through C may not overload
>>> it.
>>>
>>> I suppose you could always back up server A on C, and then include that
>>> backup when C backs up on B. Complicated but less time consuming than
>>> reinstalling A.
>>>
>>> RB wrote:
>>> > I did try 1 network card (the backup server is built-in and no other
>>> > server
>>> > is having this problem) and connected the servers on another switch in
>>> > the
>>> > stack. I don't have two extra cards so at the moment can't try to
>>> > connect
>>> > them directly, and I can't use the current connections because these
>>> > servers
>>> > are critical to the network. I also don't think it's a Backup Exec
>>> > issue as
>>> > 1) it only runs 1/2 hour a day and 2) the connection is dropped all
>>> > throughout the day.
>>> >
>>> > I am trying to tweak log on permissions in both servers and on the DC
>>> > in the
>>> > policies that would apply. It may be a log-in locally right
>>> > misconfiguration but I've just done that so only time will tell now.
>>> >
>>> > Thanks for the suggestions. I may try a reverse backup just to see
>>> > what
>>> > happens.
>>> >
>>> >
>>> > "Greg O" <(E-Mail Removed)> wrote in message
>>> > news:(E-Mail Removed) oups.com...
>>> > > Maybe you could try backing up in the opposite direction. Instead of
>>> > > backing up files from Server A onto server B, try backing up files
>>> > > from
>>> > > server B onto server A. See if the connection drops out then. I
>>> > > assume
>>> > > you tried replacing the network cards involved. You could even try a
>>> > > direct crossover connection cable between them to rule out a hub or
>>> > > switch problem.
>>> > >
>>> > >
>>> > > RB wrote:
>>> > >> Here is the setup. We have 4 Windows 2000 servers at this site
>>> > >> with
>>> > >> about
>>> > >> 30 XP workstations. This is a site within a 3 site domain. There
>>> > >> is one
>>> > >> DC
>>> > >> at each site. But I'm only having a problem with this site so I
>>> > >> don't
>>> > >> think
>>> > >> the sites are the problem. One of our Win2000 servers has gobs of
>>> > >> hard
>>> > >> drive space so I set up the other 3 to backup to it, that is, they
>>> > >> are
>>> > >> just
>>> > >> backing up to files - not a tape drive. Each backup runs 2 hours
>>> > >> apart
>>> > >> and
>>> > >> only takes about a half hour so they are not overlapping. But this
>>> > >> still
>>> > >> isn't the problem as this runs smoothly. The problem is 1 server
>>> > >> is now
>>> > >> loosing it's connection to this server every couple of minutes and
>>> > >> after
>>> > >> a
>>> > >> few minutes it connects back up. Neither of these are the DC, they
>>> > >> are
>>> > >> just
>>> > >> plain servers. Nothing has changed as far as I can tell. On the
>>> > >> backup
>>> > >> server there are no errors in the logs but on the server that's
>>> > >> sending
>>> > >> the
>>> > >> file I get numerous Event ID 3034 with the last part of the data
>>> > >> being
>>> > >> c0000133 which, according to numerous sources, indicates a time
>>> > >> error of
>>> > >> some kind. I've checked the clocks and they are spot on to the DC
>>> > >> and I
>>> > >> set
>>> > >> up a policy to allow a 10 minute deviation of clocks, but like I
>>> > >> said,
>>> > >> they
>>> > >> are spot on. I've even checked the CMOS times and dates and time
>>> > >> zones
>>> > >> and
>>> > >> cannot find a single second off between the two.
>>> > >>
>>> > >> If I go into the network neighborhood on the sending server and
>>> > >> click on
>>> > >> the
>>> > >> backup server when this is happening it asks for a password and I
>>> > >> can
>>> > >> login
>>> > >> with the domain admin account and nothing else. Once I do this the
>>> > >> connection is back up. I've checked all DNS and WINS records,
>>> > >> can't find
>>> > >> a
>>> > >> name or IP conflict, cleared caches, and the 3034 error is the only
>>> > >> error
>>> > >> (warning, really) in the log of the sending server, the backup
>>> > >> server has
>>> > >> numerous (seconds apart) event ID 100 from MSFTPSVS which says "The
>>> > >> server
>>> > >> was unable to logon the Windows NT account "Administrator" due to
>>> > >> the
>>> > >> follwoing error: Logon Failure: unknown user name or bad password.
>>> > >> The
>>> > >> data
>>> > >> is the error code." with the last entry in the data being 0000052e.
>>> > >> Is
>>> > >> this
>>> > >> the sending server failing to log on? Or a service failing to
>>> > >> start on
>>> > >> the
>>> > >> backup server? The other two servers that send backups are not
>>> > >> having a
>>> > >> problem at all. Sometimes the connection will last a day or two
>>> > >> but
>>> > >> mostly
>>> > >> the error pops up every 20-30 minutes, which doesn't correspond to
>>> > >> anything
>>> > >> that I can find. No other servers anywhere are having any problem,
>>> > >> it's
>>> > >> just between these two. IT'S NOT THE CLOCK for those of you about
>>> > >> to
>>> > >> suggest to reset the time. I've done this and checked this many,
>>> > >> many
>>> > >> times. The only transfer between these two servers is the backup
>>> > >> which
>>> > >> is a
>>> > >> half-hour at the most, once a day.
>>> > >>
>>> > >> This really is a smooth-running network and I've been chasing this
>>> > >> error
>>> > >> for
>>> > >> a few months now. Any suggestions at all? Anything else I can
>>> > >> monitor?
>>> > >> Neither of these accesses the internet and yes, I've checked for
>>> > >> viruses
>>> > >> and
>>> > >> spyware.
>>> > >>
>>> > >> Thanks to anyone who got this far!
>>> > >
>>
>
>

The backup didn't run last night but I'm hoping it was something else. The
3034 error is still there after I put a new NIC in both servers, used
different cables, and plugged into different switches. Hopefully it was
something else that broke the connection last night I'll see how the backups
run this weekend. I'm at my wits end.


"randy benson" <randybenson&@twrbenson&dot&.com> wrote in message
news:(E-Mail Removed)...
> If the 3034 error doesn't go away, try other NICs and cables before going
> further afield.
>
> If goes away after giving Everyone logon locally rights, see if you still
> get good results removing Everyone and restricting it to Authenticated
> Users.
>
> HTH,
> Randy Benson
>
> "RB" <(E-Mail Removed)> wrote in message
> news:44d0b119$(E-Mail Removed)...
>> Already not using Kerberos. Can't backup to another server as only the
>> backup server has enough space on a hard drive. This may be a moot point
>> as since I gave logon locally rights to everyone it's been working
>> smoothly. The 3034 error is still popping up every couple of minutes but
>> since all the backups and file transfers are working smoothly it may be a
>> totally unrelated error. Is there any risk to leaving the logon locally
>> right to everyone? I may just need to back up now and find the specific
>> right that was needed.
>>
>> Is there a monitor or something I can set up to figure out the 3034
>> error? Again, the time is synched and spot on with all other servers.
>> I've been backwards and fowards through time settings and I can't see
>> time being the issue.
>>
>> Thanks again for the suggestions.
>>
>>
>> <(E-Mail Removed)> wrote in message
>> news:(E-Mail Removed) oups.com...
>>>
>>> Perhaps you could also not use Kerberos, as a time difference is only
>>> significant for Kerberos. Unless you are worried about someone packet
>>> sniffing on the network Kerberos might not be needed anyway.
>>>
>>> (E-Mail Removed) wrote:
>>>> Assuming the servers are on the same subnet there may be a workaround.
>>>> You could route the server traffic through one of the other servers
>>>> using RRAS. So if there are problems with connecting servers A and B
>>>> you make a static route from A through server C. To do this you shold
>>>> be able to make a static route on Server A's network card, telling it
>>>> the route to B is using Server C as a gateway. Since C works ok with A
>>>> and B (if I understand you) then this should not drop out. I know it's
>>>> not a real solution but the extra traffic through C may not overload
>>>> it.
>>>>
>>>> I suppose you could always back up server A on C, and then include that
>>>> backup when C backs up on B. Complicated but less time consuming than
>>>> reinstalling A.
>>>>
>>>> RB wrote:
>>>> > I did try 1 network card (the backup server is built-in and no other
>>>> > server
>>>> > is having this problem) and connected the servers on another switch
>>>> > in the
>>>> > stack. I don't have two extra cards so at the moment can't try to
>>>> > connect
>>>> > them directly, and I can't use the current connections because these
>>>> > servers
>>>> > are critical to the network. I also don't think it's a Backup Exec
>>>> > issue as
>>>> > 1) it only runs 1/2 hour a day and 2) the connection is dropped all
>>>> > throughout the day.
>>>> >
>>>> > I am trying to tweak log on permissions in both servers and on the DC
>>>> > in the
>>>> > policies that would apply. It may be a log-in locally right
>>>> > misconfiguration but I've just done that so only time will tell now.
>>>> >
>>>> > Thanks for the suggestions. I may try a reverse backup just to see
>>>> > what
>>>> > happens.
>>>> >
>>>> >
>>>> > "Greg O" <(E-Mail Removed)> wrote in message
>>>> > news:(E-Mail Removed) oups.com...
>>>> > > Maybe you could try backing up in the opposite direction. Instead
>>>> > > of
>>>> > > backing up files from Server A onto server B, try backing up files
>>>> > > from
>>>> > > server B onto server A. See if the connection drops out then. I
>>>> > > assume
>>>> > > you tried replacing the network cards involved. You could even try
>>>> > > a
>>>> > > direct crossover connection cable between them to rule out a hub or
>>>> > > switch problem.
>>>> > >
>>>> > >
>>>> > > RB wrote:
>>>> > >> Here is the setup. We have 4 Windows 2000 servers at this site
>>>> > >> with
>>>> > >> about
>>>> > >> 30 XP workstations. This is a site within a 3 site domain. There
>>>> > >> is one
>>>> > >> DC
>>>> > >> at each site. But I'm only having a problem with this site so I
>>>> > >> don't
>>>> > >> think
>>>> > >> the sites are the problem. One of our Win2000 servers has gobs of
>>>> > >> hard
>>>> > >> drive space so I set up the other 3 to backup to it, that is, they
>>>> > >> are
>>>> > >> just
>>>> > >> backing up to files - not a tape drive. Each backup runs 2 hours
>>>> > >> apart
>>>> > >> and
>>>> > >> only takes about a half hour so they are not overlapping. But
>>>> > >> this still
>>>> > >> isn't the problem as this runs smoothly. The problem is 1 server
>>>> > >> is now
>>>> > >> loosing it's connection to this server every couple of minutes and
>>>> > >> after
>>>> > >> a
>>>> > >> few minutes it connects back up. Neither of these are the DC,
>>>> > >> they are
>>>> > >> just
>>>> > >> plain servers. Nothing has changed as far as I can tell. On the
>>>> > >> backup
>>>> > >> server there are no errors in the logs but on the server that's
>>>> > >> sending
>>>> > >> the
>>>> > >> file I get numerous Event ID 3034 with the last part of the data
>>>> > >> being
>>>> > >> c0000133 which, according to numerous sources, indicates a time
>>>> > >> error of
>>>> > >> some kind. I've checked the clocks and they are spot on to the DC
>>>> > >> and I
>>>> > >> set
>>>> > >> up a policy to allow a 10 minute deviation of clocks, but like I
>>>> > >> said,
>>>> > >> they
>>>> > >> are spot on. I've even checked the CMOS times and dates and time
>>>> > >> zones
>>>> > >> and
>>>> > >> cannot find a single second off between the two.
>>>> > >>
>>>> > >> If I go into the network neighborhood on the sending server and
>>>> > >> click on
>>>> > >> the
>>>> > >> backup server when this is happening it asks for a password and I
>>>> > >> can
>>>> > >> login
>>>> > >> with the domain admin account and nothing else. Once I do this
>>>> > >> the
>>>> > >> connection is back up. I've checked all DNS and WINS records,
>>>> > >> can't find
>>>> > >> a
>>>> > >> name or IP conflict, cleared caches, and the 3034 error is the
>>>> > >> only error
>>>> > >> (warning, really) in the log of the sending server, the backup
>>>> > >> server has
>>>> > >> numerous (seconds apart) event ID 100 from MSFTPSVS which says
>>>> > >> "The
>>>> > >> server
>>>> > >> was unable to logon the Windows NT account "Administrator" due to
>>>> > >> the
>>>> > >> follwoing error: Logon Failure: unknown user name or bad password.
>>>> > >> The
>>>> > >> data
>>>> > >> is the error code." with the last entry in the data being
>>>> > >> 0000052e. Is
>>>> > >> this
>>>> > >> the sending server failing to log on? Or a service failing to
>>>> > >> start on
>>>> > >> the
>>>> > >> backup server? The other two servers that send backups are not
>>>> > >> having a
>>>> > >> problem at all. Sometimes the connection will last a day or two
>>>> > >> but
>>>> > >> mostly
>>>> > >> the error pops up every 20-30 minutes, which doesn't correspond to
>>>> > >> anything
>>>> > >> that I can find. No other servers anywhere are having any
>>>> > >> problem, it's
>>>> > >> just between these two. IT'S NOT THE CLOCK for those of you about
>>>> > >> to
>>>> > >> suggest to reset the time. I've done this and checked this many,
>>>> > >> many
>>>> > >> times. The only transfer between these two servers is the backup
>>>> > >> which
>>>> > >> is a
>>>> > >> half-hour at the most, once a day.
>>>> > >>
>>>> > >> This really is a smooth-running network and I've been chasing this
>>>> > >> error
>>>> > >> for
>>>> > >> a few months now. Any suggestions at all? Anything else I can
>>>> > >> monitor?
>>>> > >> Neither of these accesses the internet and yes, I've checked for
>>>> > >> viruses
>>>> > >> and
>>>> > >> spyware.
>>>> > >>
>>>> > >> Thanks to anyone who got this far!
>>>> > >
>>>
>>
>>
>
>

Perhaps you should run performance monitor on both servers with all the
relevant network settings, log the results and then you could tell us
if anything happens in the log when the connection is lost. For example
if a program or processor is using too much of the processor then the
connection might be lost. Sometimes for example RAID and mirroring
overloads the cpu with the system process and makes other processes not
run properly.

RB wrote:
> The backup didn't run last night but I'm hoping it was something else. The
> 3034 error is still there after I put a new NIC in both servers, used
> different cables, and plugged into different switches. Hopefully it was
> something else that broke the connection last night I'll see how the backups
> run this weekend. I'm at my wits end.
>
>
> "randy benson" <randybenson&@twrbenson&dot&.com> wrote in message
> news:(E-Mail Removed)...
> > If the 3034 error doesn't go away, try other NICs and cables before going
> > further afield.
> >
> > If goes away after giving Everyone logon locally rights, see if you still
> > get good results removing Everyone and restricting it to Authenticated
> > Users.
> >
> > HTH,
> > Randy Benson
> >
> > "RB" <(E-Mail Removed)> wrote in message
> > news:44d0b119$(E-Mail Removed)...
> >> Already not using Kerberos. Can't backup to another server as only the
> >> backup server has enough space on a hard drive. This may be a moot point
> >> as since I gave logon locally rights to everyone it's been working
> >> smoothly. The 3034 error is still popping up every couple of minutes but
> >> since all the backups and file transfers are working smoothly it may be a
> >> totally unrelated error. Is there any risk to leaving the logon locally
> >> right to everyone? I may just need to back up now and find the specific
> >> right that was needed.
> >>
> >> Is there a monitor or something I can set up to figure out the 3034
> >> error? Again, the time is synched and spot on with all other servers.
> >> I've been backwards and fowards through time settings and I can't see
> >> time being the issue.
> >>
> >> Thanks again for the suggestions.
> >>
> >>
> >> <(E-Mail Removed)> wrote in message
> >> news:(E-Mail Removed) oups.com...
> >>>
> >>> Perhaps you could also not use Kerberos, as a time difference is only
> >>> significant for Kerberos. Unless you are worried about someone packet
> >>> sniffing on the network Kerberos might not be needed anyway.
> >>>
> >>> (E-Mail Removed) wrote:
> >>>> Assuming the servers are on the same subnet there may be a workaround.
> >>>> You could route the server traffic through one of the other servers
> >>>> using RRAS. So if there are problems with connecting servers A and B
> >>>> you make a static route from A through server C. To do this you shold
> >>>> be able to make a static route on Server A's network card, telling it
> >>>> the route to B is using Server C as a gateway. Since C works ok with A
> >>>> and B (if I understand you) then this should not drop out. I know it's
> >>>> not a real solution but the extra traffic through C may not overload
> >>>> it.
> >>>>
> >>>> I suppose you could always back up server A on C, and then include that
> >>>> backup when C backs up on B. Complicated but less time consuming than
> >>>> reinstalling A.
> >>>>
> >>>> RB wrote:
> >>>> > I did try 1 network card (the backup server is built-in and no other
> >>>> > server
> >>>> > is having this problem) and connected the servers on another switch
> >>>> > in the
> >>>> > stack. I don't have two extra cards so at the moment can't try to
> >>>> > connect
> >>>> > them directly, and I can't use the current connections because these
> >>>> > servers
> >>>> > are critical to the network. I also don't think it's a Backup Exec
> >>>> > issue as
> >>>> > 1) it only runs 1/2 hour a day and 2) the connection is dropped all
> >>>> > throughout the day.
> >>>> >
> >>>> > I am trying to tweak log on permissions in both servers and on the DC
> >>>> > in the
> >>>> > policies that would apply. It may be a log-in locally right
> >>>> > misconfiguration but I've just done that so only time will tell now.
> >>>> >
> >>>> > Thanks for the suggestions. I may try a reverse backup just to see
> >>>> > what
> >>>> > happens.
> >>>> >
> >>>> >
> >>>> > "Greg O" <(E-Mail Removed)> wrote in message
> >>>> > news:(E-Mail Removed) oups.com...
> >>>> > > Maybe you could try backing up in the opposite direction. Instead
> >>>> > > of
> >>>> > > backing up files from Server A onto server B, try backing up files
> >>>> > > from
> >>>> > > server B onto server A. See if the connection drops out then. I
> >>>> > > assume
> >>>> > > you tried replacing the network cards involved. You could even try
> >>>> > > a
> >>>> > > direct crossover connection cable between them to rule out a hub or
> >>>> > > switch problem.
> >>>> > >
> >>>> > >
> >>>> > > RB wrote:
> >>>> > >> Here is the setup. We have 4 Windows 2000 servers at this site
> >>>> > >> with
> >>>> > >> about
> >>>> > >> 30 XP workstations. This is a site within a 3 site domain. There
> >>>> > >> is one
> >>>> > >> DC
> >>>> > >> at each site. But I'm only having a problem with this site so I
> >>>> > >> don't
> >>>> > >> think
> >>>> > >> the sites are the problem. One of our Win2000 servers has gobs of
> >>>> > >> hard
> >>>> > >> drive space so I set up the other 3 to backup to it, that is, they
> >>>> > >> are
> >>>> > >> just
> >>>> > >> backing up to files - not a tape drive. Each backup runs 2 hours
> >>>> > >> apart
> >>>> > >> and
> >>>> > >> only takes about a half hour so they are not overlapping. But
> >>>> > >> this still
> >>>> > >> isn't the problem as this runs smoothly. The problem is 1 server
> >>>> > >> is now
> >>>> > >> loosing it's connection to this server every couple of minutes and
> >>>> > >> after
> >>>> > >> a
> >>>> > >> few minutes it connects back up. Neither of these are the DC,
> >>>> > >> they are
> >>>> > >> just
> >>>> > >> plain servers. Nothing has changed as far as I can tell. On the
> >>>> > >> backup
> >>>> > >> server there are no errors in the logs but on the server that's
> >>>> > >> sending
> >>>> > >> the
> >>>> > >> file I get numerous Event ID 3034 with the last part of the data
> >>>> > >> being
> >>>> > >> c0000133 which, according to numerous sources, indicates a time
> >>>> > >> error of
> >>>> > >> some kind. I've checked the clocks and they are spot on to the DC
> >>>> > >> and I
> >>>> > >> set
> >>>> > >> up a policy to allow a 10 minute deviation of clocks, but like I
> >>>> > >> said,
> >>>> > >> they
> >>>> > >> are spot on. I've even checked the CMOS times and dates and time
> >>>> > >> zones
> >>>> > >> and
> >>>> > >> cannot find a single second off between the two.
> >>>> > >>
> >>>> > >> If I go into the network neighborhood on the sending server and
> >>>> > >> click on
> >>>> > >> the
> >>>> > >> backup server when this is happening it asks for a password and I
> >>>> > >> can
> >>>> > >> login
> >>>> > >> with the domain admin account and nothing else. Once I do this
> >>>> > >> the
> >>>> > >> connection is back up. I've checked all DNS and WINS records,
> >>>> > >> can't find
> >>>> > >> a
> >>>> > >> name or IP conflict, cleared caches, and the 3034 error is the
> >>>> > >> only error
> >>>> > >> (warning, really) in the log of the sending server, the backup
> >>>> > >> server has
> >>>> > >> numerous (seconds apart) event ID 100 from MSFTPSVS which says
> >>>> > >> "The
> >>>> > >> server
> >>>> > >> was unable to logon the Windows NT account "Administrator" due to
> >>>> > >> the
> >>>> > >> follwoing error: Logon Failure: unknown user name or bad password.
> >>>> > >> The
> >>>> > >> data
> >>>> > >> is the error code." with the last entry in the data being
> >>>> > >> 0000052e. Is
> >>>> > >> this
> >>>> > >> the sending server failing to log on? Or a service failing to
> >>>> > >> start on
> >>>> > >> the
> >>>> > >> backup server? The other two servers that send backups are not
> >>>> > >> having a
> >>>> > >> problem at all. Sometimes the connection will last a day or two
> >>>> > >> but
> >>>> > >> mostly
> >>>> > >> the error pops up every 20-30 minutes, which doesn't correspond to
> >>>> > >> anything
> >>>> > >> that I can find. No other servers anywhere are having any
> >>>> > >> problem, it's
> >>>> > >> just between these two. IT'S NOT THE CLOCK for those of you about
> >>>> > >> to
> >>>> > >> suggest to reset the time. I've done this and checked this many,
> >>>> > >> many
> >>>> > >> times. The only transfer between these two servers is the backup
> >>>> > >> which
> >>>> > >> is a
> >>>> > >> half-hour at the most, once a day.
> >>>> > >>
> >>>> > >> This really is a smooth-running network and I've been chasing this
> >>>> > >> error
> >>>> > >> for
> >>>> > >> a few months now. Any suggestions at all? Anything else I can
> >>>> > >> monitor?
> >>>> > >> Neither of these accesses the internet and yes, I've checked for
> >>>> > >> viruses
> >>>> > >> and
> >>>> > >> spyware.
> >>>> > >>
> >>>> > >> Thanks to anyone who got this far!
> >>>> > >
> >>>
> >>
> >>
> >
> >