sp_helpexternlogin 12.5.0.1 --> 12.5.0.3

Hi,

we are at the point of upgrading to 12.5.0.3, and I wonder whether a bug 
we are experiencing has been fixed in this new version.

We use DirectConnect ( to Oracle), and sp_addexternlogin to specify the 
userid/passwd to be used at the remote end. These settings are per 
Sybase login and per remote destination, and we have two or three dozen, 
as shown by sp_helpexternlogin.  Trouble is, every now and then all 
these settings disappear from the system, and we have to re-enter them 
manually. We can exclude foul play by now, i.e. no one has come to 
delete them.

We are on Solaris, and I suspect we are being hit by one of the last 
manifestations of the multithreading issues there. With DirectConnect, 
there is a parallel Omni-xx process. For the externlogin, uid/pwd 
settings are stored, if I remember well, in the sysattributes table in 
the master database. The parallel process needs to read the table, in 
order to find out how to connect. I can imagine some contention 
happening there, but all this is just guesswork.

Has anybody else experienced this bug ? Is it fixed in 12.5.0.3 ?

Andrew Schonberger

0
Andrew
2/20/2003 10:07:40 AM
sybase.gateways 470 articles. 0 followers. Follow

3 Replies
1057 Views

Similar Articles

[PageSpeed] 13
Get it on Google Play
Get it on Apple App Store

Andrew Schonberger wrote:

> we are at the point of upgrading to 12.5.0.3, and I wonder whether a bug
> we are experiencing has been fixed in this new version.
> 
> We use DirectConnect ( to Oracle), and sp_addexternlogin to specify the
> userid/passwd to be used at the remote end. These settings are per
> Sybase login and per remote destination, and we have two or three dozen,
> as shown by sp_helpexternlogin.  Trouble is, every now and then all
> these settings disappear from the system, and we have to re-enter them
> manually. We can exclude foul play by now, i.e. no one has come to
> delete them.

	How do you know for certain? If they are being deleted (or
	perhaps the table is being truncated), something must be
	doing it. If you can recognise a pattern to when it happens,
	perhaps it might provide a clue. You could also consider
	auditing the relevant system tables and look at permissions
	(I don't think you can revoke permissions on system tables
	but its easy enough to check). A grep thru the installmaster
	script might also be helpful (in case there's a bug somewhere
	that deletes them in another system sproc).

> We are on Solaris, and I suspect we are being hit by one of the last
> manifestations of the multithreading issues there. With DirectConnect,
> there is a parallel Omni-xx process. For the externlogin, uid/pwd
> settings are stored, if I remember well, in the sysattributes table in
> the master database. The parallel process needs to read the table, in
> order to find out how to connect. I can imagine some contention
> happening there, but all this is just guesswork.
> 
> Has anybody else experienced this bug ? Is it fixed in 12.5.0.3 ?

	If its not mentioned in the cover letter have you tried asking
	Sybase Tech Support?

-am	� 2003
0
Anthony
2/20/2003 9:04:15 PM
--------------090301000902080905050102
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 8bit

Anthony,

thanks for taking this up. We have found it is a known bug in the Sybase 
problem database:

      http://info.sybase.com/resolution/detail.stm?id_number=10808059
    Problem Description:
    During upgrade to 12.5 the external login rows in sysattributes that
    pertain to external logins to DirectConnect, other ASEs,etc. are
    lost (or deleted) by the upgrade process! These are the rows where
    object_type = 'EL'.
    Resolution:
    CR: 269430
    Title sp_droplogin will also drop ALL external login lines from
    sysattributes Workaround
    Sanitized
    The sp_droplogin stored procedure will drop all external login rows
    from sysattributes regardless of qualified user id.

Andrew


Anthony Mandic wrote:

>Andrew Schonberger wrote:
>
>>we are at the point of upgrading to 12.5.0.3, and I wonder whether a bug
>>we are experiencing has been fixed in this new version.
>>
>>We use DirectConnect ( to Oracle), and sp_addexternlogin to specify the
>>userid/passwd to be used at the remote end. These settings are per
>>Sybase login and per remote destination, and we have two or three dozen,
>>as shown by sp_helpexternlogin.  Trouble is, every now and then all
>>these settings disappear from the system, and we have to re-enter them
>>manually. We can exclude foul play by now, i.e. no one has come to
>>delete them.
>>
>
>	How do you know for certain? If they are being deleted (or
>	perhaps the table is being truncated), something must be
>	doing it. If you can recognise a pattern to when it happens,
>	perhaps it might provide a clue. You could also consider
>	auditing the relevant system tables and look at permissions
>	(I don't think you can revoke permissions on system tables
>	but its easy enough to check). A grep thru the installmaster
>	script might also be helpful (in case there's a bug somewhere
>	that deletes them in another system sproc).
>
>>We are on Solaris, and I suspect we are being hit by one of the last
>>manifestations of the multithreading issues there. With DirectConnect,
>>there is a parallel Omni-xx process. For the externlogin, uid/pwd
>>settings are stored, if I remember well, in the sysattributes table in
>>the master database. The parallel process needs to read the table, in
>>order to find out how to connect. I can imagine some contention
>>happening there, but all this is just guesswork.
>>
>>Has anybody else experienced this bug ? Is it fixed in 12.5.0.3 ?
>>
>
>	If its not mentioned in the cover letter have you tried asking
>	Sybase Tech Support?
>
>-am	� 2003
>


--------------090301000902080905050102
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<html>
<head>
</head>
<body>
Anthony,<br>
<br>
thanks for taking this up. We have found it is a known bug in the Sybase
problem database:<br>
<blockquote>&nbsp; <font size="2" face="Arial"><a href="http://info.sybase.com/resolution/detail.stm?id_number=10808059" target="_blank">
http://info.sybase.com/resolution/detail.stm?id_number=10808059</a>
  </font><br>
  <b><font size="2" face="Arial">Problem Description:</font></b><br>
  <font size="2" face="Arial">During upgrade to 12.5 the external login rows 
in sysattributes that pertain to external logins to DirectConnect, other ASEs,etc.
are lost (or deleted) by the upgrade process! These are the rows where object_type
= 'EL'.</font><br>
  <b><font size="2" face="Arial">Resolution:</font></b><br>
  <font color="#ff0000" size="2" face="Arial">CR: 269430</font><br>
  <font color="#ff0000" size="2" face="Arial"> Title sp_droplogin will also
drop ALL external login lines from sysattributes Workaround</font><br>
  <font color="#ff0000" size="2" face="Arial"> Sanitized</font><br>
  <font color="#ff0000" size="2" face="Arial"> The sp_droplogin stored procedure
will drop all external login rows from sysattributes regardless of qualified
user id.</font><br>
  </blockquote>
Andrew<br>
  <br>
  <br>
Anthony Mandic wrote:<br>
  <blockquote type="cite" cite="mid:3E5542CF.460C9A00@start.com.au">
    <pre wrap="">Andrew Schonberger wrote:<br><br></pre>
    <blockquote type="cite">
      <pre wrap="">we are at the point of upgrading to 12.5.0.3, and I wonder whether a bug<br>we are experiencing has been fixed in this new version.<br><br>We use DirectConnect ( to Oracle), and sp_addexternlogin to specify the<br>userid/passwd to be used at the remote end. These settings are per<br>Sybase login and per remote destination, and we have two or three dozen,<br>as shown by sp_helpexternlogin.  Trouble is, every now and then all<br>these settings disappear from the system, and we have to re-enter them<br>manually. We can exclude foul play by now, i.e. no one has come to<br>delete them.<br></pre>
      </blockquote>
      <pre wrap=""><!----><br>	How do you know for certain? If they are being deleted (or<br>	perhaps the table is being truncated), something must be<br>	doing it. If you can recognise a pattern to when it happens,<br>	perhaps it might provide a clue. You could also consider<br>	auditing the relevant system tables and look at permissions<br>	(I don't think you can revoke permissions on system tables<br>	but its easy enough to check). A grep thru the installmaster<br>	script might also be helpful (in case there's a bug somewhere<br>	that deletes them in another system sproc).<br><br></pre>
      <blockquote type="cite">
        <pre wrap="">We are on Solaris, and I suspect we are being hit by one of the last<br>manifestations of the multithreading issues there. With DirectConnect,<br>there is a parallel Omni-xx process. For the externlogin, uid/pwd<br>settings are stored, if I remember well, in the sysattributes table in<br>the master database. The parallel process needs to read the table, in<br>order to find out how to connect. I can imagine some contention<br>happening there, but all this is just guesswork.<br><br>Has anybody else experienced this bug ? Is it fixed in 12.5.0.3 ?<br></pre>
        </blockquote>
        <pre wrap=""><!----><br>	If its not mentioned in the cover letter have you tried asking<br>	Sybase Tech Support?<br><br>-am	&copy; 2003<br></pre>
        </blockquote>
        <br>
        </body>
        </html>

--------------090301000902080905050102--

0
Andrew
2/21/2003 9:24:44 AM
Andrew Schonberger wrote:

> We have found it is a known bug in the Sybase problem database:
> 
> http://info.sybase.com/resolution/detail.stm?id_number=10808059

	The link says its related to upgrades. So that would only cover
	in place upgrades and use of the online database command after
	a load. Are either of these used by you when you see it happening?

-am	� 2003
0
Anthony
2/22/2003 4:58:47 AM
Reply:

Similar Artilces:

Which Version 12.5.0.3, 12.5.0.4, 12.5.1
We are currrently running 11.9.2 on an old HPUX server (development). We are getting a newer HPUX server and our DBA is asking us which version we should install. It will be a clean install. Are there any major problems we may encounter when the 11.9.2 databases are installed/upgraded to any of the above version. We have numerous stored procs, views and triggers. Once all applications are fully tested, we will be upgrading the production server to whichever version we choose. I hope that made sense. TIA Scott ...

12.5.0.1 to 12.5.0.3 upgrade
We are planning to upgrade from 12.5.0.1 to 12.5.0.3 on Win NT After the upgradation the Sybase service did not start. The error encountered was: "Error 1053: The service did not respond to the start or control request in timely fashion" We reverted back to the previous setup and the service started and we could connect to the server. What could be wrong?? Perhaps you could post the Sybase errorlog. You should be able to find it under %SYBASE%\%SYBASE_ASE%\install\errorlog, unless it was given a different name during installation. <Mihir> wrote in message...

Testday - Friday, May18th: Firefox 1.5.0.12
Hi Mozilla Community Testers!, Please join the Mozilla QA community for Testday on Friday, May 18th, 2007! This week, we will be looking at the Firefox 1.5.0.12 and 2.0.0.4 and Thunderbird 1.5.0.12 Release Candidates The event will run Friday, May 18th, 2007, from 7am - 5pm PDT Please visit http://wiki.mozilla.org/Mozilla_QA_Community:Firefox_2004rc3/15012_Community_Test_Day for all the info regarding the event. Jump into the #testday channel on irc.mozilla.org on Friday to participate. We hope to see you there! Thanks. Carsten Mozilla QA ...

Limited options on AIX going from 12.5.0.1 to 12.5.0.3
We find the upgrade options on AIX 4.3.3 and ASE 12.5.0.1 quite limiting and frankly, just want to hear if anyone else feels such frustration. Since ASE 12.5.0.3 64-bit requires OS levels of AIX 5.1, we must combine the OS upgrade with the ASE binary upgrade in one event. Then there is no easy ASE version rollback plan, without rolling back OS. Woo is me. USAdba wrote: > We find the upgrade options on AIX 4.3.3 and ASE 12.5.0.1 quite limiting and > frankly, just want to hear if anyone else feels such frustration. Since ASE > 12.5.0.3 64-bit requires OS levels of AIX 5...

Smoke [5.12.0] v5.12.0-RC5-1-g675ac12 FAIL(XF) netbsd 5.0.2 (i386/1 cpu)
Automated smoke report for 5.12.0 patch 675ac12c19e6fe00eff6e604a7d637bf621997ef v5.12.0-RC5-1-g675ac12 p5netbsd: Intel 686-class (i386/1 cpu) on netbsd - 5.0.2 using cc version 4.1.3 20080704 prerelease (NetBSD nb2 20081120) smoketime 6 hours 36 minutes (average 49 minutes 36 seconds) Summary: FAIL(XF) O = OK F = Failure(s), extended report at the bottom X = Failure(s) under TEST but not under harness ? = still running or test results not (yet) available Build failures during: - = unknown or N/A c = Configure, m = make, M = make (after miniperl), t ...

Advice for copying from win2000 Syb 12.5.0.3 to Linux 12.5.0.3
I need to find a mechanism for copying DBs from my producction Windows 2000 Server: Adaptive Server Enterprise/12.5.0.3/EBF 10695 IR/P/NT (IX86)/OS 4.0/rel12503/1915/32-bit/OPT/Wed Jan 22 23:28:11 2003Adaptive Server Enterprise/12.5.0.3/EBF 10695 IR/P/NT (IX86)/OS 4.0/rel12503/1915/32-bit/OPT/Wed Jan 22 23:28:11 2003 To my test environment: Adaptive Server Enterprise/12.5.0.3/EBF 11340 ESD#3/P/Linux Intel/Linux 2.4.18-18.7.xsmp i686/rel12503/1933/32-bit/OPT/Thu Jul 31 23:09:54 2003 I understand I cannot simply dump from windows and load into Linux. Any other suggestions? thank...

Migrating 12.5.0 GA to 12.5.0.1
I installed the generally available release on solaris 8. I then went to back up and got a an error about block zero being an invalid block. Looking at the web, this was a known problem on partitions > 4GB in size, fixed in 12.5.0..1 I then went and download EBF 10700 (interum release) . It was not clear to me what EBF I should download. This release did not contain loc OCS libraries for 12.5.0.1. So I then have downloaded EBF 10423. BTW: I gunzip'ed 10700 and gnu un tared and got a directory check sum error?? Have looked over sybase site and am still unclear how it...

A Possible fix for Suse 8.0 and ASE 12.5.0.2 (upgrade from 12.5.0.1)
Hi All, I had a similar problem to what a lot of you were experiencing... with regards to glibc issues with ASE and Suse 8.0, and the server not starting. After a lot of reading up, and running stack traces, and various other things to try and get my ASE 12.5.0.1 esd#2 upgraded to 12.5.0.2 esd#1 (and of course ready for when 12.5.0.3 comes out), I came to the conclusion that the 2.2.5-55 and 2.2.5-158 were both not suitable for running 12.5.0.2 (my 12.5.0.1 worked well tho). I found the following article from SuSe... ftp://ftp.is.co.za/linux/distributions/suse/8.0/unsorted...

Smoke [5.12.3] v5.12.3-RC3-1-gc024706 FAIL(c) netbsd 5.0.2 (i386/1 cpu)
Automated smoke report for 5.12.3 patch c0247066c34077145eaa8f33eddcbb64fa3b0f8d v5.12.3-RC3-1-gc024706 qemu386netbsd5-0-2.bandsman.co.uk: AMD 686-class (i386/1 cpu) on netbsd - 5.0.2 using ? unknown cc version smoketime 3 seconds (average 3 seconds) Summary: FAIL(c) O = OK F = Failure(s), extended report at the bottom X = Failure(s) under TEST but not under harness ? = still running or test results not (yet) available Build failures during: - = unknown or N/A c = Configure, m = make, M = make (after miniperl), t = make test-prep v5.12.3-RC3-1-gc...

Upgrading from 12.5.0.3 to 12.5.1
Has anyone experienced any major problems when upgrading from ASE 12.5.0.3 to 12.5.1? Thanks Mike No, we have been running with since it was released <mike> wrote in message news:40200e3f.3cdd.846930886@sybase.com... > Has anyone experienced any major problems when upgrading > from ASE 12.5.0.3 to 12.5.1? > > Thanks > Mike mike, if you're doing binary overlay (the same as ebf installation), make sure that none of the databases is set to "read only" mode. otherwise, you'll get 806 when ase tries to dump tran with truncate_only o...

Upgrading from 12.5.0.3 to 12.5.1
I was about to apply 32-bit 12.5.1 ESD#1 ( EBF 11559 ) to an existing installation of 12-bit 12.5.0.3 ESD#3 on 64-bit Solaris8 .... .... when I saw that the EBF 11559 / ESD#1 for 12.5.1 consists of subdirectories from JVM, I realised that it was an installer ( instead of the usual thing of overlaying the binaries on the existing installation with the new binaries from the ESD ). And, reading the EBF_11559_Readme.txt, it specifically said: If this shows that the ASE Server is NOT at version 12.5.1, stop here. Do NOT proceed. Your software needs to be at version ...

Upgrading from 12.5.0.3 to 12.5.1
Has anyone experienced any major problems when upgrading from ASE 12.5.0.3 to 12.5.1? Thanks Mike If your close to max memory for the amount that you have configured then you may want to ensure that there is sufficient space for the 12.5.1 ase binary. It pushed total logical memory over max memory when i upgraded. > Has anyone experienced any major problems when upgrading > from ASE 12.5.0.3 to 12.5.1? > > Thanks > Mike Hello, How to upgrade from 12.5.03 to 12.5.1? Is there a CD or EBF that can be downloaded? We are still using 11.9.2 for production a...

stalled or no download: Mozilla/5.0 (X11; U; Linux i686; en-GB; rv:1.8.0.3) Gecko/20060425 SUSE/1.5.0.3-7 Firefox/1.5.0.3
I experience stalled or no download of all (not only specific web-sites) using firefox while another browser "Konquereor" is working normally. Firefox is looking up.. connecting... transfering and then stalled sometimes I get the message "I cant find the server", suprisingly because the status shows transfering, or a login window for the router shows up what has not been seen with the other web browser. A second Linux PC in the network is working as supposed, the router is wide open... connection settings are alredy checked. May be anybody has a hint? Ber...

stalled transfer Mozilla/5.0 (X11; U; Linux i686; en-GB; rv:1.8.0.3) Gecko/20060425 SUSE/1.5.0.3-7 Firefox/1.5.0.3
I experience stalled or no download of all (not only specific web-sites) using firefox while another browser "Konquereor" is working normally. Firefox is looking up.. connecting... transferring and then stalled sometimes I get the message "I cant find the server", surprisingly because the status shows transferring, or a login window for the router shows up what has not been seen with the other web browser. A second Linux PC in the network is working as supposed, the router is wide open... connection settings are already checked. May be anybody has a hint? ...

Smoke [5.12.0] v5.12.0-RC3-4-g4dfa489 FAIL(XF) netbsd 5.0.2 (i386/1 cpu)
Automated smoke report for 5.12.0 patch 4dfa489a8b653b67eac56af0c26b7caa9b97b904 v5.12.0-RC3-4-g4dfa489 p5netbsd: Intel 686-class (i386/1 cpu) on netbsd - 5.0.2 using cc version 4.1.3 20080704 prerelease (NetBSD nb2 20081120) smoketime 6 hours 35 minutes (average 49 minutes 30 seconds) Summary: FAIL(XF) O = OK F = Failure(s), extended report at the bottom X = Failure(s) under TEST but not under harness ? = still running or test results not (yet) available Build failures during: - = unknown or N/A c = Configure, m = make, M = make (after miniperl), t ...

Web resources about - sp_helpexternlogin 12.5.0.1 --> 12.5.0.3 - sybase.gateways

Resources last updated: 11/22/2015 9:27:10 PM