[Veritas-bu] Issues with NetBackup after applying Solaris patches

This is a multi-part message in MIME format.


------_=_NextPart_001_01C6597F.67A453CA
Content-Type: text/plain;
        charset="US-ASCII"
Content-Transfer-Encoding: quoted-printable

Hey all!=0D=0A=0D=0AWe've begun rolling out our annual patch cluster for So=
laris (we're=0D=0Arunning 8 right now) and as soon as it was installed, we =
began to=0D=0Aexperience some very strange issues with NetBackup=2E=0D=0A=
=0D=0AEnvironment:=0D=0A=0D=0ASolaris 8 (patch cluster installed was 10/200=
5)=2E We have also since=0D=0Apatched the CE card and TCP to the latest ava=
ilable patch levels=0D=0ASun F15k domain (4x8, single board)=0D=0A4 IP addr=
esses across 4 subnets on both the master/media and the media=0D=0Aserver=
=0D=0ANetBackup 5=2E0 MP6 for Solaris=0D=0ANetBackup 5=2E0 MP5 for NDMP=0D=
=0AFilers are:=0D=0A    NetApp  w/tape attached is Release 6=2E5=2E2R1P10       
=0D=
=0A     NetApp w/out tape are Release 6=2E4=2E5=0D=0A=0D=0AIssues:=0D=0A=0D=0A1=
=2E  All NDMP backups with DAR enabled no longer work=2E However, if we set=
=0D=0Athe HIST_FILE =3D N in the policy include configuration, it works fin=
e=2E 3=0D=0ANetApps, same problem with each one=2E 1 has tape attached thru=
 the SAN=0D=0Aand the others back up thru it=2E=0D=0A=0D=0A2=2E  211, 213 a=
nd 195 errors happen nightly on clients=2E Backups running=0D=0Athru the ma=
ster/media itself do not experience issues, however, backups=0D=0Arunning t=
hru media servers or SAN media servers will get these errors=2E=0D=0AThe ba=
ckups will generally re-run successfully, if they run thru another=0D=0Amed=
ia server=2E SAN Media server backups will fail completely=2E This=0D=0Adoe=
sn't happen on the same clients every night and some nights doesn't=0D=0Aha=
ppen at all=2E=0D=0A=0D=0A3=2E  Backups get sent to the worklist but the PI=
D dies so quickly that=0D=0Athe job looks to be running (bpdbjobs) but the =
PID isn't active on the=0D=0Amaster anymore, nor is there any real activity=
 within Netbackup (no=0D=0Atapes mounted, not writing going on, no database=
 entries being created,=0D=0Aetc=2E)=2E The only way to get rid of these *a=
ctive* jobs is to completely=0D=0Arecycle Netbackup (bp=2Ekill_all)=2E=0D=
=0A=0D=0ASteps taken so far:=0D=0A=0D=0AWe have disabled all but the PRIMAR=
Y interface on the Master and Media=0D=0Aserver and are no longer getting 2=
11, 213 and 195 errors, however the=0D=0ANDMP/w DAR still do not work=2E We=
're going to start adding interfaces=0D=0Aback in and see at what point it =
breaks again=2E =0D=0A=0D=0AIf anyone has seen anything like this can you p=
lease let me know if=0D=0Ayou've resolved it? We're at backline w/ Sun and =
about to be escalated=0D=0Ato backline w/ Veritas=2E  =0D=0A=0D=0AThanks!=
=0D=0AKate=0D=0A=0D=0A=0D=0A=0D=0A=0D=0A-----------------------------------=
------=0D=0AThis e-mail may contain confidential or privileged information=
=2E  If=0D=0Ayou=0D=0Athink you have received this e-mail in error, please =
advise the=0D=0Asender by=0D=0Areply e-mail and then delete this e-mail imm=
ediately=2E  Thank you=2E=0D=0AAetna=0D=0A
------_=_NextPart_001_01C6597F.67A453CA
Content-Type: text/html;
        charset="US-ASCII"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3=2E2//EN">=0D=0A<HTML>=0D=0A<HEAD>=
=0D=0A<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; charset=3Dus-=
ascii">=0D=0A<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version=
 6=2E5=2E7233=2E28">=0D=0A<TITLE>Issues with NetBackup after applying Solar=
is patches</TITLE>=0D=0A</HEAD>=0D=0A<BODY>=0D=0A<!-- Converted from text/r=
tf format -->=0D=0A=0D=0A<P><FONT SIZE=3D2 FACE=3D"Georgia">Hey all!</FONT>=
=0D=0A</P>=0D=0A=0D=0A<P><FONT SIZE=3D2 FACE=3D"Georgia">We've begun rollin=
g out our annual patch cluster for Solaris (we're running 8 right now) and =
as soon as it was installed, we began to experience some very strange issue=
s with NetBackup=2E</FONT></P>=0D=0A=0D=0A<P><FONT SIZE=3D2 FACE=3D"Georgia=
">Environment:</FONT>=0D=0A</P>=0D=0A=0D=0A<P><FONT SIZE=3D2 FACE=3D"Georgi=
a">Solaris 8 (patch cluster installed was 10/2005)=2E We have also since pa=
tched the CE card and TCP to the latest available patch levels</FONT></P>=
=0D=0A=0D=0A<P><FONT SIZE=3D2 FACE=3D"Georgia">Sun F15k domain (4x8, single=
 board)</FONT>=0D=0A=0D=0A<BR><FONT SIZE=3D2 FACE=3D"Georgia">4 IP addresse=
s across 4 subnets on both the master/media and the media server</FONT>=0D=
=0A=0D=0A<BR><FONT SIZE=3D2 FACE=3D"Georgia">NetBackup 5=2E0 MP6 for Solari=
s</FONT>=0D=0A=0D=0A<BR><FONT SIZE=3D2 FACE=3D"Georgia">NetBackup 5=2E0 MP5=
 for NDMP</FONT>=0D=0A=0D=0A<BR><FONT SIZE=3D2 FACE=3D"Georgia">Filers are:=
</FONT>=0D=0A=0D=0A<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; <FONT SIZ=
E=3D2 FACE=3D"Georgia">NetApp&nbsp; w/tape attached is Release 6=2E5=2E2R1P=
10&nbsp;&nbsp; </FONT>=0D=0A=0D=0A<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp; <FONT SIZE=3D2 FACE=3D"Georgia">NetApp w/out tape are Release 6=2E4=
=2E5</FONT>=0D=0A</P>=0D=0A=0D=0A<P><FONT SIZE=3D2 FACE=3D"Georgia">Issues:=
</FONT>=0D=0A</P>=0D=0A=0D=0A<P><FONT SIZE=3D2 FACE=3D"Georgia">1=2E&nbsp; =
All NDMP backups with DAR enabled no longer work=2E However, if we set the =
HIST_FILE =3D N in the policy include configuration, it works fine=2E 3 Net=
Apps, same problem with each one=2E 1 has tape attached thru the SAN and th=
e others back up thru it=2E</FONT></P>=0D=0A=0D=0A<P><FONT SIZE=3D2 FACE=3D=
"Georgia">2=2E&nbsp; 211, 213 and 195 errors happen nightly on clients=2E B=
ackups running thru the master/media itself do not experience issues, howev=
er, backups running thru media servers or SAN media servers will get these =
errors=2E The backups will generally re-run successfully, if they run thru =
another media server=2E SAN Media server backups will fail completely=2E Th=
is doesn't happen on the same clients every night and some nights doesn't h=
appen at all=2E</FONT></P>=0D=0A=0D=0A<P><FONT SIZE=3D2 FACE=3D"Georgia">3=
=2E&nbsp; Backups get sent to the worklist but the PID dies so quickly that=
 the job looks to be running (bpdbjobs) but the PID isn't active on the mas=
ter anymore, nor is there any real activity within Netbackup (no tapes moun=
ted, not writing going on, no database entries being created, etc=2E)=2E Th=
e only way to get rid of these *active* jobs is to completely recycle Netba=
ckup (bp=2Ekill_all)=2E</FONT></P>=0D=0A=0D=0A<P><FONT SIZE=3D2 FACE=3D"Geo=
rgia">Steps taken so far:</FONT>=0D=0A</P>=0D=0A=0D=0A<P><FONT SIZE=3D2 FAC=
E=3D"Georgia">We have disabled all but the PRIMARY interface on the Master =
and Media server and are no longer getting 211, 213 and 195 errors, however=
 the NDMP/w DAR still do not work=2E We're going to start adding interfaces=
 back in and see at what point it breaks again=2E </FONT></P>=0D=0A=0D=0A<P=
><FONT SIZE=3D2 FACE=3D"Georgia">If anyone has seen anything like this can =
you please let me know if you've resolved it? We're at backline w/ Sun and =
about to be escalated to backline w/ Veritas=2E&nbsp; </FONT></P>=0D=0A=0D=
=0A<P><FONT SIZE=3D2 FACE=3D"Georgia">Thanks!<BR>=0D=0AKate</FONT>=0D=0A</P=
>=0D=0A=0D=0A</BODY>=0D=0A</HTML>=0D=0A=0D=0A<HTML><BODY><P><hr size=3D1></=
P><br>=0D=0A<P><STRONG><br>=0D=0AThis e-mail may contain confidential or pr=
ivileged information=2E  If<br>=0D=0Ayou<br>=0D=0Athink you have received t=
his e-mail in error, please advise the<br>=0D=0Asender by<br>=0D=0Areply e-=
mail and then delete this e-mail immediately=2E  Thank you=2E<br>=0D=0AAetn=
a<br>=0D=0A</STRONG></P></BODY></HTML>=0D=0A
------_=_NextPart_001_01C6597F.67A453CA--