Sybase NNTP forums - End Of Life (EOL)

The NNTP forums from Sybase - forums.sybase.com - are now closed.

All new questions should be directed to the appropriate forum at the SAP Community Network (SCN).

Individual products have links to the respective forums on SCN, or you can go to SCN and search for your product in the search box (upper right corner) to find your specific developer center.

DDLGEN panic (yes, green thread)

7 posts in Cluster Last posting was on 2010-05-20 23:39:21.0Z
Whipple Posted on 2009-05-22 16:07:15.0Z
Sender: 23c0.4a16ca32.1804289383@sybase.com
From: Whipple
Newsgroups: sybase.public.ase.cluster
Subject: DDLGEN panic (yes, green thread)
X-Mailer: WebNews to Mail Gateway v1.1t
Message-ID: <4a16cdb3.243e.1681692777@sybase.com>
NNTP-Posting-Host: 10.22.241.41
X-Original-NNTP-Posting-Host: 10.22.241.41
Date: 22 May 2009 09:07:15 -0700
X-Trace: forums-1-dub 1243008435 10.22.241.41 (22 May 2009 09:07:15 -0700)
X-Original-Trace: 22 May 2009 09:07:15 -0700, 10.22.241.41
Lines: 9
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.cluster:138
Article PK: 48421

Sybase 12_5 on Solaris 8 with java 1.4.2_6.
Starting May 17, the ../sybcent32/ddlgen utility script
panics with the "libthread in green" message.

I swear to you that nothing was changed, upgraded, or
patched before the 17th. All the .so libs are in the correct
place, and JAVA_HOME is set correctly. Other ideas why this
would work one day, but not the next, with regard to the
absence of changes to anything?


Jeff Tallman [Sybase] Posted on 2009-05-23 01:23:02.0Z
From: "Jeff Tallman [Sybase]" <jeff.tallman@sybase.com>
User-Agent: Thunderbird 2.0.0.21 (Windows/20090302)
MIME-Version: 1.0
Newsgroups: sybase.public.ase.cluster
Subject: Re: DDLGEN panic (yes, green thread)
References: <4a16cdb3.243e.1681692777@sybase.com>
In-Reply-To: <4a16cdb3.243e.1681692777@sybase.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
NNTP-Posting-Host: vip152.sybase.com
X-Original-NNTP-Posting-Host: vip152.sybase.com
Message-ID: <4a174ff6@forums-1-dub>
Date: 22 May 2009 18:23:02 -0700
X-Trace: forums-1-dub 1243041782 10.22.241.152 (22 May 2009 18:23:02 -0700)
X-Original-Trace: 22 May 2009 18:23:02 -0700, vip152.sybase.com
Lines: 23
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.cluster:139
Article PK: 48419

I think support for jdk/jre 1.4.2 was eol'd by Sun some time ago.... I
do know we had to do a lot of repackaging due to a jdk/jre EOL...could
be that the EOL expired....technically never heard of a "bomb" in it and
assuming nothing was changed on the OS, it should have kept working
though. One thing I can think of is maybe on one day the . SYBASE.sh
was in a different order than a later program and it affected $LIBPATH
or something. I am assuming you tried a clean shell??

Jeff Tallman
Enterprise Data Management Products Technical Evangelism
jeff.tallman@sybase.com
http://blogs.sybase.com/database

Whipple wrote:
> Sybase 12_5 on Solaris 8 with java 1.4.2_6.
> Starting May 17, the ../sybcent32/ddlgen utility script
> panics with the "libthread in green" message.
>
> I swear to you that nothing was changed, upgraded, or
> patched before the 17th. All the .so libs are in the correct
> place, and JAVA_HOME is set correctly. Other ideas why this
> would work one day, but not the next, with regard to the
> absence of changes to anything?


Whipple Posted on 2009-05-25 14:26:04.0Z
Sender: 2b38.4a1aa6e6.1804289383@sybase.com
From: Whipple
Newsgroups: sybase.public.ase.cluster
Subject: Re: DDLGEN panic (yes, green thread)
X-Mailer: WebNews to Mail Gateway v1.1t
Message-ID: <4a1aaa7c.2baa.1681692777@sybase.com>
References: <4a174ff6@forums-1-dub>
NNTP-Posting-Host: 10.22.241.41
X-Original-NNTP-Posting-Host: 10.22.241.41
Date: 25 May 2009 07:26:04 -0700
X-Trace: forums-1-dub 1243261564 10.22.241.41 (25 May 2009 07:26:04 -0700)
X-Original-Trace: 25 May 2009 07:26:04 -0700, 10.22.241.41
Lines: 38
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.cluster:140
Article PK: 48424

Thanks for the info, Jeff. It's a customer's site for whom I
provide first support (I'm not the regular admin).

I tried again today and ddlgen seems to be working again. At
least it didn't panic; I didn't do the full extraction.

New shell? Well, yes, insofar as ending my telnet session
and starting a new one. But not changing from Korn to
something else.

I'll keep looking.

> I think support for jdk/jre 1.4.2 was eol'd by Sun some
> time ago.... I do know we had to do a lot of repackaging
> due to a jdk/jre EOL...could be that the EOL
> expired....technically never heard of a "bomb" in it and
> assuming nothing was changed on the OS, it should have
> kept working though. One thing I can think of is maybe
> on one day the . SYBASE.sh was in a different order than
> a later program and it affected $LIBPATH or something. I
> am assuming you tried a clean shell??
>
> Jeff Tallman
> Enterprise Data Management Products Technical Evangelism
> jeff.tallman@sybase.com
> http://blogs.sybase.com/database
>
> Whipple wrote:
> > Sybase 12_5 on Solaris 8 with java 1.4.2_6.
> > Starting May 17, the ../sybcent32/ddlgen utility script
> > panics with the "libthread in green" message.
> >
> > I swear to you that nothing was changed, upgraded, or
> > patched before the 17th. All the .so libs are in the
> > correct place, and JAVA_HOME is set correctly. Other
> > ideas why this would work one day, but not the next,
> > with regard to the absence of changes to anything?


"Mark A. Parsons" <iron_horse Posted on 2009-05-25 14:48:26.0Z
From: "Mark A. Parsons" <iron_horse@no_spamola.compuserve.com>
User-Agent: Thunderbird 1.5.0.10 (Windows/20070221)
MIME-Version: 1.0
Newsgroups: sybase.public.ase.cluster
Subject: Re: DDLGEN panic (yes, green thread)
References: <4a174ff6@forums-1-dub> <4a1aaa7c.2baa.1681692777@sybase.com>
In-Reply-To: <4a1aaa7c.2baa.1681692777@sybase.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-Antivirus: avast! (VPS 090523-0, 05/23/2009), Outbound message
X-Antivirus-Status: Clean
NNTP-Posting-Host: vip152.sybase.com
X-Original-NNTP-Posting-Host: vip152.sybase.com
Message-ID: <4a1aafba@forums-1-dub>
Date: 25 May 2009 07:48:26 -0700
X-Trace: forums-1-dub 1243262906 10.22.241.152 (25 May 2009 07:48:26 -0700)
X-Original-Trace: 25 May 2009 07:48:26 -0700, vip152.sybase.com
Lines: 68
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.cluster:141
Article PK: 48423

I think what Jeff was suggesting was that on the day in question (when ddlgen didn't work) you may have executed some
programs/scripts in such an order to cause $LIBPATH (or another environment variable?) to be reset in a way that caused
the issue.

Consider:

Some programs will throw away current settings by resetting environment variables, eg, LIBPATH=/my/new/lib/path =>
discards any values previously assigned to LIBPATH. This could cause a problem if follow-on programs are expecting to
find a file in the old LIBPATH settings, but instead find a similarly named file in /my/new/lib/path (or they don't find
the desired file at all).

Some programs will append to the beginning of already-set environment variables, eg,
LIBPATH=/my/new/lib/path:${LIBPATH}. This could cause a problem if follow-on programs are expecting to find a file in
the old ${LIBPATH} setting, but instead find a similarly named file in /my/new/lib/path.

Some programs will append to the end of already-set environment variables, eg, LIBPATH=${LIBPATH}:/my/new/lib/path.
This could cause a problem if follow-on programs are expecting to find a file in /my/new/lib/path, but instead find a
similarly named file in the old ${LIBPATH} setting.

-----------

So, you may want to compare environment variable settings on 'good' and 'bad' days to see if there are any differences.

Alternatively, before using ddlgen make sure you source a resource file which will put all ddlgen required paths at the
beginning of their respective environment variables. (or modify ddlgen to (re)set the environment variables for you)

Whipple wrote:
> Thanks for the info, Jeff. It's a customer's site for whom I
> provide first support (I'm not the regular admin).
>
> I tried again today and ddlgen seems to be working again. At
> least it didn't panic; I didn't do the full extraction.
>
> New shell? Well, yes, insofar as ending my telnet session
> and starting a new one. But not changing from Korn to
> something else.
>
> I'll keep looking.
>
>
>> I think support for jdk/jre 1.4.2 was eol'd by Sun some
>> time ago.... I do know we had to do a lot of repackaging
>> due to a jdk/jre EOL...could be that the EOL
>> expired....technically never heard of a "bomb" in it and
>> assuming nothing was changed on the OS, it should have
>> kept working though. One thing I can think of is maybe
>> on one day the . SYBASE.sh was in a different order than
>> a later program and it affected $LIBPATH or something. I
>> am assuming you tried a clean shell??
>>
>> Jeff Tallman
>> Enterprise Data Management Products Technical Evangelism
>> jeff.tallman@sybase.com
>> http://blogs.sybase.com/database
>>
>> Whipple wrote:
>>> Sybase 12_5 on Solaris 8 with java 1.4.2_6.
>>> Starting May 17, the ../sybcent32/ddlgen utility script
>>> panics with the "libthread in green" message.
>>>
>>> I swear to you that nothing was changed, upgraded, or
>>> patched before the 17th. All the .so libs are in the
>>> correct place, and JAVA_HOME is set correctly. Other
>>> ideas why this would work one day, but not the next,
>>> with regard to the absence of changes to anything?


"Mark A. Parsons" <iron_horse Posted on 2009-05-25 14:56:19.0Z
From: "Mark A. Parsons" <iron_horse@no_spamola.compuserve.com>
User-Agent: Thunderbird 1.5.0.10 (Windows/20070221)
MIME-Version: 1.0
Newsgroups: sybase.public.ase.cluster
Subject: Re: DDLGEN panic (yes, green thread)
References: <4a174ff6@forums-1-dub> <4a1aaa7c.2baa.1681692777@sybase.com> <4a1aafba@forums-1-dub>
In-Reply-To: <4a1aafba@forums-1-dub>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-Antivirus: avast! (VPS 090523-0, 05/23/2009), Outbound message
X-Antivirus-Status: Clean
NNTP-Posting-Host: vip152.sybase.com
X-Original-NNTP-Posting-Host: vip152.sybase.com
Message-ID: <4a1ab193$1@forums-1-dub>
Date: 25 May 2009 07:56:19 -0700
X-Trace: forums-1-dub 1243263379 10.22.241.152 (25 May 2009 07:56:19 -0700)
X-Original-Trace: 25 May 2009 07:56:19 -0700, vip152.sybase.com
Lines: 86
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.cluster:142
Article PK: 48422

'course, there's also the possibility that someone was modifying various resource files that you would normally access
during your rounds (eg, global resource files accessed during the login process) which in turn could have caused you
problems.

I've seen folks make changes to resource files thinking that they were only affecting their login, but in reality caused
havoc for other users/processes running on the same machine.

Mark A. Parsons wrote:
> I think what Jeff was suggesting was that on the day in question (when
> ddlgen didn't work) you may have executed some programs/scripts in such
> an order to cause $LIBPATH (or another environment variable?) to be
> reset in a way that caused the issue.
>
> Consider:
>
> Some programs will throw away current settings by resetting environment
> variables, eg, LIBPATH=/my/new/lib/path => discards any values
> previously assigned to LIBPATH. This could cause a problem if follow-on
> programs are expecting to find a file in the old LIBPATH settings, but
> instead find a similarly named file in /my/new/lib/path (or they don't
> find the desired file at all).
>
> Some programs will append to the beginning of already-set environment
> variables, eg, LIBPATH=/my/new/lib/path:${LIBPATH}. This could cause a
> problem if follow-on programs are expecting to find a file in the old
> ${LIBPATH} setting, but instead find a similarly named file in
> /my/new/lib/path.
>
> Some programs will append to the end of already-set environment
> variables, eg, LIBPATH=${LIBPATH}:/my/new/lib/path. This could cause a
> problem if follow-on programs are expecting to find a file in
> /my/new/lib/path, but instead find a similarly named file in the old
> ${LIBPATH} setting.
>
> -----------
>
> So, you may want to compare environment variable settings on 'good' and
> 'bad' days to see if there are any differences.
>
> Alternatively, before using ddlgen make sure you source a resource file
> which will put all ddlgen required paths at the beginning of their
> respective environment variables. (or modify ddlgen to (re)set the
> environment variables for you)
>
>
>
>
> Whipple wrote:
>> Thanks for the info, Jeff. It's a customer's site for whom I
>> provide first support (I'm not the regular admin).
>>
>> I tried again today and ddlgen seems to be working again. At
>> least it didn't panic; I didn't do the full extraction.
>>
>> New shell? Well, yes, insofar as ending my telnet session
>> and starting a new one. But not changing from Korn to
>> something else.
>>
>> I'll keep looking.
>>
>>
>>> I think support for jdk/jre 1.4.2 was eol'd by Sun some
>>> time ago.... I do know we had to do a lot of repackaging
>>> due to a jdk/jre EOL...could be that the EOL
>>> expired....technically never heard of a "bomb" in it and
>>> assuming nothing was changed on the OS, it should have
>>> kept working though. One thing I can think of is maybe
>>> on one day the . SYBASE.sh was in a different order than
>>> a later program and it affected $LIBPATH or something. I
>>> am assuming you tried a clean shell??
>>>
>>> Jeff Tallman
>>> Enterprise Data Management Products Technical Evangelism
>>> jeff.tallman@sybase.com
>>> http://blogs.sybase.com/database
>>>
>>> Whipple wrote:
>>>> Sybase 12_5 on Solaris 8 with java 1.4.2_6.
>>>> Starting May 17, the ../sybcent32/ddlgen utility script
>>>> panics with the "libthread in green" message.
>>>>
>>>> I swear to you that nothing was changed, upgraded, or
>>>> patched before the 17th. All the .so libs are in the
>>>> correct place, and JAVA_HOME is set correctly. Other
>>>> ideas why this would work one day, but not the next,
>>>> with regard to the absence of changes to anything?


Whipple Posted on 2009-06-08 13:58:07.0Z
Sender: 4019.4a2d16d6.1804289383@sybase.com
From: whipple
Newsgroups: sybase.public.ase.cluster
Subject: Re: DDLGEN panic (yes, green thread)
X-Mailer: WebNews to Mail Gateway v1.1t
Message-ID: <4a2d18ef.406c.1681692777@sybase.com>
References: <4a1ab193$1@forums-1-dub>
NNTP-Posting-Host: 10.22.241.41
X-Original-NNTP-Posting-Host: 10.22.241.41
Date: 8 Jun 2009 06:58:07 -0700
X-Trace: forums-1-dub 1244469487 10.22.241.41 (8 Jun 2009 06:58:07 -0700)
X-Original-Trace: 8 Jun 2009 06:58:07 -0700, 10.22.241.41
Lines: 105
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.ase.cluster:151
Article PK: 48428

Thank you (back from vacation now).
I didn't analyze LIBPATH in exactly the way you describe,
but will watch for it (was a primary suspect, of course). I
compared env vars in my shell (process owner) against the
env files that the process owner sources, but found no
differences. I do understand, however, the subtleties you
describe. BTW--ddlgen suddenly stopped panicking, and didn't
panic during my vacay. yippee.
--mw

> 'course, there's also the possibility that someone was
> modifying various resource files that you would normally
> access during your rounds (eg, global resource files
> accessed during the login process) which in turn could
> have caused you problems.
>
> I've seen folks make changes to resource files thinking
> that they were only affecting their login, but in reality
> caused havoc for other users/processes running on the
> same machine.
>
> Mark A. Parsons wrote:
> > I think what Jeff was suggesting was that on the day in
> > question (when ddlgen didn't work) you may have
> > executed some programs/scripts in such an order to
> > cause $LIBPATH (or another environment variable?) to be
> > reset in a way that caused the issue.
> > Consider:
> >
> > Some programs will throw away current settings by
> > resetting environment variables, eg,
> > LIBPATH=/my/new/lib/path => discards any values
> previously assigned to LIBPATH. This could cause a
> > problem if follow-on programs are expecting to find a
> > file in the old LIBPATH settings, but instead find a
> > similarly named file in /my/new/lib/path (or they don't
> > find the desired file at all).
> > Some programs will append to the beginning of
> > already-set environment variables, eg,
> > LIBPATH=/my/new/lib/path:${LIBPATH}. This could cause a
> problem if follow-on programs are expecting to find a
> > file in the old ${LIBPATH} setting, but instead find a
> > similarly named file in /my/new/lib/path.
> >
> > Some programs will append to the end of already-set
> > environment variables, eg,
> > LIBPATH=${LIBPATH}:/my/new/lib/path. This could cause a
> problem if follow-on programs are expecting to find a file
> > in /my/new/lib/path, but instead find a similarly named
> > file in the old ${LIBPATH} setting.
> >
> > -----------
> >
> > So, you may want to compare environment variable
> > settings on 'good' and 'bad' days to see if there are
> > any differences.
> > Alternatively, before using ddlgen make sure you source
> > a resource file which will put all ddlgen required
> > paths at the beginning of their respective environment
> > variables. (or modify ddlgen to (re)set the
> > environment variables for you)
> >
> >
> >
> > Whipple wrote:
> >> Thanks for the info, Jeff. It's a customer's site for
> whom I >> provide first support (I'm not the regular
> admin). >>
> >> I tried again today and ddlgen seems to be working
> again. At >> least it didn't panic; I didn't do the full
> extraction. >>
> >> New shell? Well, yes, insofar as ending my telnet
> session >> and starting a new one. But not changing from
> Korn to >> something else.
> >>
> >> I'll keep looking.
> >>
> >>
> >>> I think support for jdk/jre 1.4.2 was eol'd by Sun
> some >>> time ago.... I do know we had to do a lot of
> repackaging >>> due to a jdk/jre EOL...could be that the
> EOL >>> expired....technically never heard of a "bomb" in
> it and >>> assuming nothing was changed on the OS, it
> should have >>> kept working though. One thing I can
> think of is maybe >>> on one day the . SYBASE.sh was in a
> different order than >>> a later program and it affected
> $LIBPATH or something. I >>> am assuming you tried a
> clean shell?? >>>
> >>> Jeff Tallman
> >>> Enterprise Data Management Products Technical
> Evangelism >>> jeff.tallman@sybase.com
> >>> http://blogs.sybase.com/database
> >>>
> >>> Whipple wrote:
> >>>> Sybase 12_5 on Solaris 8 with java 1.4.2_6.
> >>>> Starting May 17, the ../sybcent32/ddlgen utility
> script >>>> panics with the "libthread in green" message.
> >>>>
> >>>> I swear to you that nothing was changed, upgraded, or
> >>>> patched before the 17th. All the .so libs are in the
> >>>> correct place, and JAVA_HOME is set correctly. Other
> >>>> ideas why this would work one day, but not the next,
> >>>> with regard to the absence of changes to anything?


vdhara Posted on 2010-05-20 23:39:21.0Z
Sender: 7a32.4bf5c7cb.1804289383@sybase.com
From: vdhara
Newsgroups: sybase.public.ase.cluster
Subject: Re: DDLGEN panic (yes, green thread)
X-Mailer: WebNews to Mail Gateway v1.1t
Message-ID: <4bf5c822.7a6b.1681692777@sybase.com>
References: <4a1aaa7c.2baa.1681692777@sybase.com>
X-Original-NNTP-Posting-Host: forums-test-vm.sybase.com
X-Original-Trace: 20 May 2010 16:39:14 -0800, forums-test-vm.sybase.com
Lines: 38
X-Original-NNTP-Posting-Host: forums-test-vm.sybase.com
X-Original-Trace: 20 May 2010 16:39:17 -0700, forums-test-vm.sybase.com
NNTP-Posting-Host: forums-master.sybase.com
X-Original-NNTP-Posting-Host: forums-master.sybase.com
Date: 20 May 2010 16:39:21 -0700
X-Trace: forums-1-dub 1274398761 10.22.108.75 (20 May 2010 16:39:21 -0700)
X-Original-Trace: 20 May 2010 16:39:21 -0700, forums-master.sybase.com
X-Authenticated-User: ngsysop
Path: forums-1-dub!forums-test-vm.sybase.com!not-for-mail
Xref: forums-1-dub sybase.public.ase.cluster:195
Article PK: 48473


> Thanks for the info, Jeff. It's a customer's site for whom
> I provide first support (I'm not the regular admin).
>
> I tried again today and ddlgen seems to be working again.
> At least it didn't panic; I didn't do the full extraction.
>
> New shell? Well, yes, insofar as ending my telnet session
> and starting a new one. But not changing from Korn to
> something else.
>
> I'll keep looking.
>
>
> > I think support for jdk/jre 1.4.2 was eol'd by Sun some
> > time ago.... I do know we had to do a lot of
> > repackaging due to a jdk/jre EOL...could be that the
> > EOL expired....technically never heard of a "bomb" in it
> > and assuming nothing was changed on the OS, it should
> > have kept working though. One thing I can think of is
> > maybe on one day the . SYBASE.sh was in a different
> > order than a later program and it affected $LIBPATH or
> > something. I am assuming you tried a clean shell??
> >
> > Jeff Tallman
> > Enterprise Data Management Products Technical Evangelism
> > jeff.tallman@sybase.com
> > http://blogs.sybase.com/database
> >
> > Whipple wrote:
> > > Sybase 12_5 on Solaris 8 with java 1.4.2_6.
> > > Starting May 17, the ../sybcent32/ddlgen utility
> > > script panics with the "libthread in green" message.
> > >
> > > I swear to you that nothing was changed, upgraded, or
> > > patched before the 17th. All the .so libs are in the
> > > correct place, and JAVA_HOME is set correctly. Other
> > > ideas why this would work one day, but not the next,
> > > with regard to the absence of changes to anything?