Hi.
We're not a huge DB2 shop and I never been near TSAfM. I'm sure it works though.
/Gus
________________________________
From: [email protected] [mailto:[email protected]] On Behalf Of Jay Slay
Sent: den 14 december 2012 14:49
To: Discussion list for Tivoli product and Tivoli Ready products.
Cc: <[email protected]>
Subject: Re: SV: [TME10] ITM 6.2.3 FP2 is FTO totally broken? [SOLVED]
I don't disagree that this should all be in the itm product by default...you are 100% correct.
despite that...if you want to get as close to your goal as possible now...db2 HADR (which is supported by IBM via db2 support) works like a charm for teps db replication.
Also, since itm supports tivoli system automation for multiplatforms, you can probably play with configuring the teps to use a VIP for the monitoring server which you should be able to set up a decent health check for. How well that reconnects during a failover, I dont know, but I would assume its similar to a network outage as far as teps is concerned. Then you can set the kfw_interface_cnps_host variable to set the virtual host name that clients use for tep.
Personally, I don't bother too much with all this, I simply start my backup teps when needed. However, I do want to give it more of an effort next year if time permits.
Finally...Remember, TSAfM is free as a limited license for itm if you own itm.
-JKS
On Dec 14, 2012, at 8:25, <[email protected]<mailto:[email protected]>> wrote:
Uhhmmm not really. I could but I would have to replicate teps DB which is not supported by IBM. Useless if you think about it. What I'm aiming at is bring TEPS H/A enabled and "fail" together with acting HUB
Currently there's no way for the LB to pick up which TEPS is connected to the acting HUB. Unless the non acting one is offline but then how do you bring it online when HUB fails to each other. I'm sure you could build some fancy situations around this, So the TEPS would have to respond with an URL stating it is not "acting" which could be picked up by the LB which could redirect accordingly.
Youre right about the DB2 part though forgot about it. Haven't used DB2 for years.....
1. TEPS must support DB replication in the same way as HUB does in FTO.
2. TEPS must if configured as FTO following acting HUB and reply with different URL than the normal login page to clients. (LB's most likely)
3. World still ends 21/12 and it's still friday (-:
/Gus
________________________________
From: [email protected]<mailto:[email protected]> [mailto:[email protected]] On Behalf Of Jay Slay
Sent: den 14 december 2012 14:09
To: Discussion list for Tivoli product and Tivoli Ready products.
Cc: <[email protected]<mailto:[email protected]>>
Subject: Re: SV: [TME10] ITM 6.2.3 FP2 is FTO totally broken? [SOLVED]
You can use a load balanced VIP for tep now, and if you use db2 for the database, there is no db license to pay for. Db2 HADR will replicate your teps db for you too.
-JKS
On Dec 14, 2012, at 7:39, <[email protected]<mailto:[email protected]>> wrote:
Hi.
Oh, yes. No need to replicate DB's or sync TEP DB changes on two TEPS'es.
But then the TEPS would have to provide H/A by itself. Either by cluster or some sort of D/R failover VMware cold standby or similar.
Personally I think IBM should drop TEPS DB support for anything but derby. Then deliver derby replication as a TEPS feature. And also provide some sort of Load balancing support for TEP clients.
I.e hit LB (round robin) with the client. The TEPS who is connected to the acting TEMS would answer correct, hence LB redirects client to the TEPS connected to the acting HUB all the time.
This would save us $ in DB license, and provide full H/A for TEPS. Would require LB infrastructure ofc. but this is propably common these days.
Then again, whatever world ends on 21/12 anyways and it's Friday (-:
/Gus
________________________________
From: [email protected]<mailto:[email protected]> [mailto:[email protected]] On Behalf Of John Alvord
Sent: den 15 november 2012 17:50
To: Discussion list for Tivoli product and Tivoli Ready products.
Subject: RE: SV: [TME10] ITM 6.2.3 FP2 is FTO totally broken? [SOLVED]
Even better would be an RFE so a TEPS can be aware of both hub TEMS and switch to the primary automatically.
Tivoli Request For Enhancement (RFE)
http://www.ibm.com/support/docview.wss?uid=swg21449404 [announcement]
http://www.ibm.com/developerworks/rfe/tivoli/
Curiously enough, the previous user interface CMW at Candle - Candle Management Workstation - had exactly that capability. No idea why it wasn't carried forwarded into CNPS which became TEPS.
Regards,
John Alvord - Ph: 1-720-396-2788 Cell: none
Customer Support - Tivoli Software - jalvord-r/[email protected]<mailto:[email protected]>
Advisory Engineer - Tivoli Monitoring - ITM infrastructure
Follow us on Twitter! @Tivolisupport<http://twitter.com/TivoliSupport> and Facebook<https://www.facebook.com/IBMTivoliSupport>
Personalize your support needs with the IBM Support Portal<http://www.ibm.com/support/electronicsupport/>
Use Service Request<http://www.ibm.com/software/support/servicerequest/quick_start.html> to get assistance!
For emails regarding a PMR, copy [email protected]<mailto:[email protected]>
For secure browser uploads: https://www.ecurep.ibm.com/app/upload
For Customer Support guidance: IBM Software Support Handbook<http://www14.software.ibm.com/webapp/set2/sas/f/handbook/home.html>
<http://twitter.com/TivoliSupport>
From: <[email protected]<mailto:[email protected]>>
To: <[email protected]<mailto:[email protected]>>,
Date: 11/15/2012 12:02 AM
Subject: RE: SV: [TME10] ITM 6.2.3 FP2 is FTO totally broken? [SOLVED]
Sent by: [email protected]<mailto:[email protected]>
________________________________
Hi. Thx John.
This is getting interesting. When I promote my secondary to acting and run tacmd listsit, tacmd listsystemlist I can see all my missing sit's and MSL's!
Next I restarted my secondary TEPS and pulled a new client from port 1920 (just to be sure) and Voila! Now I can see everything.
Not sure why I needed to restart secondary TEPS. Seems like the TEPS can not pull all data from the TEMS when it is in "mirror" mode.
I'll just have to remember restarting TEPS'es whenever a HUB becomes acting.
Maybe it's a good idea to open an Enhancement req.To signal TEPS to reread TEMS data whenever a HUB becomes acting?
/Gus
________________________________
From: [email protected]<mailto:[email protected]> [mailto:[email protected]] On Behalf Of John Alvord
Sent: den 15 november 2012 00:06
To: Discussion list for Tivoli product and Tivoli Ready products.
Subject: Re: SV: [TME10] ITM 6.2.3 FP2 is FTO totally broken?
A number of tables should be identical between the two hub TEMSes
TEMS Database Tables Synchronized with FTO Configuration
https://www.ibm.com/support/docview.wss?uid=swg21503175
including the TSITDESC and TNAME tables which are needed for situations.
However, I cannot vouch that a TEPS connected to a hub TEMS/backup role will show identical results. a hub TEMS in backup role isn't doing a lot of things. A tacmd listsit should be accurate plus or minus a few seconds of updates. There are also cases of extreme overload, where the hub TEMS gets REALLY busy - sometimes for many minutes at a time and sync'ing can be delayed.
There are lots of rarer cases of breakage where things go wrong, but pretty rare.
Regards,
John Alvord - Ph: 1-720-396-2788 Cell: none
Customer Support - Tivoli Software - jalvord-r/[email protected]<mailto:[email protected]>
Advisory Engineer - Tivoli Monitoring - ITM infrastructure
Follow us on Twitter! @Tivolisupport<http://twitter.com/TivoliSupport> and Facebook<https://www.facebook.com/IBMTivoliSupport>
Personalize your support needs with the IBM Support Portal<http://www.ibm.com/support/electronicsupport/>
Use Service Request<http://www.ibm.com/software/support/servicerequest/quick_start.html> to get assistance!
For emails regarding a PMR, copy [email protected]<mailto:[email protected]>
For secure browser uploads: https://www.ecurep.ibm.com/app/upload
For Customer Support guidance: IBM Software Support Handbook<http://www14.software.ibm.com/webapp/set2/sas/f/handbook/home.html>
<http://twitter.com/TivoliSupport>
From: <[email protected]<mailto:[email protected]>>
To: <[email protected]<mailto:[email protected]>>,
Date: 11/14/2012 10:49 AM
Subject: SV: [TME10] ITM 6.2.3 FP2 is FTO totally broken?
Sent by: [email protected]<mailto:[email protected]>
________________________________
Hi.
Yes seems so. Just for the record. will/should all normal situations be visible from the mirror TEPS?
I would assume so but can't be sure.
/Gus
________________________________
Från: [email protected]<mailto:[email protected]> [[email protected]<mailto:[email protected]>] för John Alvord [jalvord-r/[email protected]<mailto:jalvord-r/[email protected]>]
Skickat: den 14 november 2012 16:07
Till: Discussion list for Tivoli product and Tivoli Ready products.
Ämne: RE: [TME10] ITM 6.2.3 FP2 is FTO totally broken?
Depending on how your testing is goes... a PMR with a view toward an APAR would be the logical next step...
Regards,
John Alvord - Ph: 1-720-396-2788 Cell: none
Customer Support - Tivoli Software - jalvord-r/[email protected]<mailto:[email protected]><mailto:jalvord-r/[email protected]>
Advisory Engineer - Tivoli Monitoring - ITM infrastructure
Follow us on Twitter! @Tivolisupport<http://twitter.com/TivoliSupport> and Facebook<https://www.facebook.com/IBMTivoliSupport>
Personalize your support needs with the IBM Support Portal<http://www.ibm.com/support/electronicsupport/>
Use Service Request<http://www.ibm.com/software/support/servicerequest/quick_start.html> to get assistance!
For emails regarding a PMR, copy [email protected]<mailto:[email protected]><mailto:[email protected]>
For secure browser uploads: https://www.ecurep.ibm.com/app/upload
For Customer Support guidance: IBM Software Support Handbook<http://www14.software.ibm.com/webapp/set2/sas/f/handbook/home.html>
<http://twitter.com/TivoliSupport>
From: <[email protected]<mailto:[email protected]>>
To: <[email protected]<mailto:[email protected]>>,
Date: 11/14/2012 06:12 AM
Subject: RE: [TME10] ITM 6.2.3 FP2 is FTO totally broken?
Sent by: [email protected]<mailto:[email protected]>
________________________________
Hi no I havent repointed my Primary TEPS to the mirror HUB. Tried this in the past on MS SQL TEPS almost always ended up in all TEPS info gone, users, groups, sit associations etc.
But it does not make sense since all sits, msl's etc runs on TEMS'es. Besides. I got one sit replicated to the mirror and seen on the mirror TEPS when the sit had the WPA agent as subscriber. But as soon as I create MSL's or distribute sit's to MSL's they do not get replicated.
I think the default heartbeat is 5 sec. And the mirror pulls the data from acting. So I would expect to see the new created MSL's in a matter of minutes. But they dont show up, ever.
Would be really interesting if you could log into your TEPS on the backup and "check things out"
If you are running on Linux. Would you dare apply FP2?
Just tried changing protocol on FTO configuration from both ip.spipe and ip.pipe to only ip.pipe. No difference.
Thing is I could live with FP1 if it wasnt for the fact that I need to do massexports of sits from one itm6 instance to the other. And atm. EIF information is lost.
/Gus
________________________________
From: [email protected]<mailto:[email protected]> [mailto:[email protected]] On Behalf Of Jay Slay
Sent: den 14 november 2012 14:44
To: Discussion list for Tivoli product and Tivoli Ready products.
Subject: Re: [TME10] ITM 6.2.3 FP2 is FTO totally broken?
Gus,
I have never logged my secondary TEPS into my backup HUB when it was acting as MIRROR. Actually, I don't think I knew that worked. I use db2 HADR, and simply start the backup TEPS only in the case of a failover.
To answer your question though, I do run 6.2.3 FP1 in both lab, test and production...so if you don't make any headway, I could try to reproduce your behavior if you like. Out of curiosity, have you repointed your primary TEPS to the backup HUB after failover? Same results?
On Nov 14, 2012, at 3:28 AM, [email protected]<mailto:[email protected]><mailto:[email protected]> wrote:
Hi
Ok I will clarify what is working and what is not.
HUBA=primary HUBB=secondary
WPAA, is running on same box as HUBA it is configured to connects to HUBA first and secondary to HUBB
WPAB, is running on same box as HUBB it is configured to connects to HUBA first and secondary to HUBB
TEPS is installed and running derby on both boxes.
Now, when HUBA is acting and HUBB is online. I create a WPS sit and distribute it directly to both WPAA and WPAB.
Then I log in to both TEP, and can se that the new sit has benn replicated to HUBB and I can se status that it is running from both TEP'S.
I switch roles of HUBA and HUBB by downing HUBA and wait for HUBB to become acting as seen in its UMC "FTO promoted HUBB as the acting HUB"
Then I check the sit from TEP on HUBB, and it is still running fine.
Then I reverse the TEMS'es and wait for HUBA to become acting again and HUBB to become mirror, checking the *ms* log for Sychronization complete. Going to sleep messages. Now I create a MSL on HUBA under WPA called TEST_WPA_FTO, put both WPA's in this MSL. Create a new WPA sit called TEST_WPA, and distributes to TEST_WPA_FTO.
Now things go bad. Neither the new sit TEST_WPA nor MSL TEST_WPS_FTO is being synchronized to to HUBB who is mirror. Even tried restarting TEP on HUBB just to be sure. Next test I did was to create a sit on HUBA, under I5/OS (just picked one) called AAA, without distribution (grey). Wating for sync complete and check for this sit on the TEP on HUBB, it's not there.
Then I switched acting again to se if anything changed, but cant find TEST_WPA_FTO, TEST_WPA nor AAA
Next I mailed list.
Looks pretty serious to me?
Either its FP2 or it's RHEL6. I have used FTO a few years ago on Solaris to sync in a new HTEMS box. It worked just fine then.
Is anyone runnig FTO on RHEL on 6.2.3 GA or FP1?
/Gus
________________________________
From: [email protected]<mailto:[email protected]><mailto:[email protected]> [[email protected]<mailto:[email protected]><mailto:[email protected]>] On Behalf Of John Alvord
Sent: den 13 november 2012 16:33
To: Discussion list for Tivoli product and Tivoli Ready products.
Subject: Re: [TME10] ITM 6.2.3 FP2 is FTO totally broken?
I am really confused.
Warehouse Proxy Agents are agents like any other. They are made part of a system generated MSL that represents all WPAs. There are a few situations which run at the WPAs - reporting error conditions etc.
Also, in my experience - each WPA should be configured to a single hub TEMS... not to both hub TEMS. The reason is that the hub TEMS with the backup role will *not* accept any agent connections. [Actually it does accept a connection and then immediately sends a "go find another TEMS and disconnects".
1) But I cant get WHP MSL's to sync and no situations subscribed to MSL's.
==> What is meant by sync'ing the WHP MSL's. ANd what does "no situations subscribed to MSLs" mean - a sentence fragement it looks like.
2) WHP situations that have agents as distribution target seems to sync just fine though.
==> WHP situations should be distributed only to WPAs, of course.
In any case... WPAs connecting to a hub TEMS should only be configured to that one hub TEMS. I suggest testing that setup out.
Regards,
John Alvord - Ph: 1-720-396-2788 Cell: none
Customer Support - Tivoli Software - jalvord-r/[email protected]<mailto:[email protected]><mailto:jalvord-r/[email protected]>
Advisory Engineer - Tivoli Monitoring - ITM infrastructure
Follow us on Twitter! @Tivolisupport<http://twitter.com/TivoliSupport> and Facebook<https://www.facebook.com/IBMTivoliSupport>
Personalize your support needs with the IBM Support Portal<http://www.ibm.com/support/electronicsupport/>
Use Service Request<http://www.ibm.com/software/support/servicerequest/quick_start.html> to get assistance!
For emails regarding a PMR, copy [email protected]<mailto:[email protected]><mailto:[email protected]>
For secure browser uploads: https://www.ecurep.ibm.com/app/upload
For Customer Support guidance: IBM Software Support Handbook<http://www14.software.ibm.com/webapp/set2/sas/f/handbook/home.html>
<http://twitter.com/TivoliSupport>
From: <[email protected]<mailto:[email protected]><mailto:[email protected]>>
To: <[email protected]<mailto:[email protected]><mailto:[email protected]>>,
Date: 11/13/2012 06:00 AM
Subject: [TME10] ITM 6.2.3 FP2 is FTO totally broken?
Sent by: [email protected]<mailto:[email protected]><mailto:[email protected]>
________________________________
Hi all.
Just installed two HTEMS'es with FTO on RHEL6 with 6.2.3 FP02 in our test env.
Both servers are identical with TEMS/TEPS on Derby/WHP all at 6.2.3 FP02. Both TEPS connect locally to each TEMS
Both WHP agents has the primary HTEMS as its primary connection and the secondary HTEMS as secondary, all by the book if you reda the manual.
The FTO "seems" to be working fine switching back and forth when I stop/start the TEMS'esr
But I cant get WHP MSL's to sync and no situations subscribed to MSL's.
WHP situations that have agents as distribution target seems to sync just fine though.
Also tried creating a sit when the secondary was acting HUB, but it never got synced to the other TEMS.
Anyone or is it just me?
/Gus
_______________________________________________
TME10 mailing list
[email protected]<mailto:[email protected]><mailto:[email protected]>
Unsubscribe:[email protected]<mailto:[email protected]><mailto:[email protected]>
_______________________________________________
TME10 mailing list
[email protected]<mailto:[email protected]><mailto:[email protected]>
Unsubscribe:[email protected]<mailto:[email protected]>
_______________________________________________
TME10 mailing list
[email protected]<mailto:[email protected]>
Unsubscribe:[email protected]<mailto:[email protected]>
_______________________________________________
TME10 mailing list
[email protected]<mailto:[email protected]>
Unsubscribe:[email protected]<mailto:[email protected]>
_______________________________________________
TME10 mailing list
[email protected]<mailto:[email protected]>
Unsubscribe:[email protected]<mailto:[email protected]>
_______________________________________________
TME10 mailing list
[email protected]<mailto:[email protected]>
Unsubscribe:[email protected]<mailto:[email protected]>
_______________________________________________
TME10 mailing list
[email protected]<mailto:[email protected]>
Unsubscribe:[email protected]<mailto:[email protected]>
_______________________________________________
TME10 mailing list
[email protected]
Unsubscribe:[email protected]
lmpx.com only provides a reader for public news (NNTP) servers. It is not
affiliated with the servers or forums shown here and is not responsible for
the content of articles, which is written by their respective authors.