Building Connections this week and troubleshooting some errors reminded me to share the process I have adopted when dealing with IBM error messages – which is to treat them as hints that can set you on the right path but also send you badly down the wrong one.
Installing Connections itself via Installation Manager. One of steps during the install requires you to specify the DB2 server, the database names and credentials to connect to them. I click validate and it fails with error CLFRP0030E and launch error!. That points to this technote which says I left a space after the hostname for the DB2 server.
I absolutely didn’t leave a space and didn’t copy/paste. Just in case (and working on the assumption that it’s always me and not the product) I cleared it all and typed carefully again. I confirmed the hostname was correct and could be reached. I also relaunched Installation Manager and started from the beginning. No luck.
It’s at this point I have to accept the error is referring to something else and that’s all the information I’m going to get from Installation Manager. So now I move to asking myself “what if I saw no error but it just failed to connect”. Well the first answer to that is to check if the connection details, hostname, credentials etc actually work at all. Having confirmed the hostname and ports (there were no firewalls turned on or virus software), I logged into the DB2 server and checked the LCUSER account. Locked out. I unlocked the account and the install then completed.
The test server in this environment is one box with everything DB2, TDI and all the applications on it. My base WebSphere install was WAS 8.5.5 FP10 since Connections System Requirements for WebSphere 8.5.5 says FP8 and higher and I wanted to test that out. Everything installed fine right up to when I went to install Connections Surveys. That’s when I hit a 2 day brick wall. Installation Manager couldn’t connect to the Deployment manager despite it being on the same server.
Well that’s odd. Deployment manager is running. The hostname resolves. The port is listening. I try to find out what the system requirements are for Connections Surveys but for 2 days last week and through the weekend the IBM system requirements pages for Survey were down. I’m stubborn so I won’t let it go. Even the Forms Experience Builder requirements for earlier versions were down. So eventually I had to leave it and move onto the production build. The work needs completing and I was suspicious that the issue might have been installing everything on one server.
I build production across 4 servers and this time I stick with WebSphere 8.5.5 FP8 just in case. When I get to the Surveys install it goes without a hitch. So back to the test server I go. Roll back Websphere to 188.8.131.52 and then forwards to FP8 (thank you Installation Manager!). Surprise surprise Surveys installed perfectly.
So. Not an issue connecting to deployment manager or port or the server running but instead “Connections Surveys cannot install onto WebSphere 184.108.40.206 at all.
Today I was contacted urgently by a site I did an install for back in early September. The install went well and I left them several months ago with working components, but apparently about a week ago people stopped being able to login to the Community server. In fact not even the SSC could access it.
.. and yet no-one had changed anything at all. I do love a good mystery so I thought it would be useful to someone (or even just future Gab) to document what I did:
- verified if port 1533 was listening using netstat -an |find /i “1533”.
- verified there were no running AV services that could interfere with the ports.
- checked if the ST services were running, in fact only about 6 were.
- tried to start some of the services that weren’t running and they failed immediately.
- since no-one touched Sametime my next guess was a Windows update that caused a problem.
- checked the Windows networking settings hadn’t been overwritten (they had) . Although those settings shouldn’t cause the services to fail completely it was worth resetting them.
- I then added vp_trace_all=1 to the [Debug] settings in the sametime.ini which creates detailed log files in the \ibm\domino\trace directory.
- having added that I could see log files being created for every service, even the ones that wouldn’t stay started. In fact those ones recreated every couple of minutes. So the services were trying to start and failing.
- reviewing the log files I could see on things like STPlaces there was a JVM error, but I put that aside for the time being in case it was a dependency issue.
- in other logs such as STDirectory I could see broken networking errors and just before that I could see a comment about switching to TLS.
A-ha! Well, that’s new.
- checking the sametime.ini I found:
which I changed to:
My guess being an incomplete TLS configuration from the SSC. Having done that the server restarted perfectly and all services started. The SSC could then access the server with no problem.
Of course once I had spent 4hrs doing that I then found a technote on it which I never would have found before I saw the TLS entry. Here’s the technote .
Sometimes it’s a rollercoaster but so long as I get things working I’m calling that a good day. Now back to building more Connections servers.
This is my presentation from Icon UK in September which discusses making client decisions whilst keeping Domino, the best mail server in the world, as your underlying architecture
How often do you hear that the business is discussing moving mail platforms because “our users want X” where X is nothing to do with the server and everything to do with the client UI. Domino remains the best mail server available but often user dissatisfaction drives a move and that comes from being asked to use the wrong client or from a bad deployment. If you’re using Domino you have an ever expanding range of clients to choose from browsers, iNotes, Verse, Traveler with iOS integration, Android applications, POP3 and IMAP.
The full presentation is here
After spending far too long (i.e. more than 10 mins) on tech support with my Amazon Echo today I finally rang them. My problem was that it refused to acknowledge I had Amazon Prime and that TuneIn radio didn’t work so I couldn’t say “Alexa, play BBC Radio 4” for instance.
Calling Amazon UK support and I’m told they entirely messed up the UK Echo devices which ship to auto register with amazon.com. Apparently tech support have to manually change your Echo registration on their end to force it to connect to the UK site instead of the US and they are “rushed off their feet” doing that as each person calls. It should be sorted by tomorrow.
- The setup was a pain and far too confusing for anyone not technically savvy (say my mother in law)
- There is no remote control supplied although one exists they are just too cheap to supply it with the device
- They shipped it with the wrong configuration assuming there is no other country other than the US.
- You have to call them to get it manually fixed
Amazon is definitely no Apple …
I upgraded my Mac at the weekend to Sierra, which went beautifully. Fast and no problems at all. Until this morning. This morning I went to connect to a customer VPN and it has disappeared from my list of VPNs on my Mac. On further checking I realised that customer ran their VPN as PPTP and PPTP is no longer supported as a VPN option on Mac OS.
I can use 3rd party VPN software and I guess removing it was Apple’s best option – leaving it in place but disabling it would have just led me into trying to make it work not realising it was no longer supported. Still a warning pre-install of “if you upgrade the following services and applications will no longer work” would have been nice if a bit much to ask for.
Consider this your warning*
*And yes I know PPTP isn’t secure and the customer shouldn’t be using it but that one isn’t my decision.
As I went to bed last night I set the alarm early, I have a lot to do this week especially since I’ll be at Icon UK for 2 days of it and I wanted to get started early. So of course today was the day my work went out of the window and I lost 10 hrs debugging one of my own servers. Let’s back up…
This weekend I was prepping my presentations for Icon UK this Thursday. One is called “Domino In The Back, Party In The Front” so I’m going to be talking about all the client options available to you using Domino as a back end.
On Sunday I had the idea of installing IMSMO (IBM Mail Services For Microsoft Outlook) on one of my lab machines. I had a customer wanting to deploy and I wanted to try and mirror their setup, plus it meant I’d have something to demo from. The lab server was already running 9.0.1 FP6 with a SHA2 SSL certificate delivering TLS1.2. I hadn’t used any web services on it in a couple of weeks so I went ahead and added IF3 (required by IMSMO) and installed the application addin service. It actually installs as a variant of Traveler (and I’ll be blogging on that later). I completed the install and Outlook worked fine. Unfortunately it was the only HTTPS service that worked. Everything failed. By failed I mean the browser – any browser – refused to connect.
So off I went to investigate why the browsers wouldn’t connect. I started with testing via SSLLabs and that reported AN F as apparently the server was demanding SSLv3 instead of TLS 1.2 Of course just about every browser will refuse to accept a negotiation of SSLV3. But why was the server suddenly demanding it when it had never done so before?
Well 10 hrs later I’d exhausted everything I could think of:
- verified notes.ini had no additional unexpected settings
- forced Disable_SSLV3=1 even though that server had been fine serving TLS 1.2 previously
- disabled internet site documents and reproduced using web configuration
- recreated the internet site and web rule documents
- generated a new keyfile from my wildcard certificates
- uninstalled IF3
- uninstalled IMSMO including all the cleanup
- scanned for anything that could be hijacking HTTPS
- restarted and restarted and restarted http and clear cache upon cache upon cache
- bothered Darren Duke for a sanity check – I believe the words “I don’t know what the hell is going on” came up
- uninstalled Domino (around hour 8) because I couldn’t spend any more time troubleshooting
After uninstalling Domino. Reinstalling up to FP6, copying in the databases and templates and restarting. I was back with TLS 1.2 again and suddenly SSLLabs was giving me an A+.
Of course then I did what I should have done in the first place (saving time is never a time saver), I built a new lab server purely for IMSMO. Installed FP6 and IF3 and the addin and everything worked perfectly including TLS1.2.
I have no idea what part of the IMSMO install , the addin or IF3, conflicted with my existing lab server configuration or what it did to force the server to only serve SSLV3 no matter how I tried to push it otherwise – but an uninstall and clean install ended up being my only fix in the time I had. Someone somewhere knows the setting that made it do that. I’d love to know what.
Now it’s 4.15am and I’m back where I thought I was at 11pm Sunday night. The 4 days work I had to fit in 2 days , I have to fit in 1 day. This week’s lesson. Never start something new when you barely have time to get the existing things completed.
See you at Icon UK
After serveral weeks travelling around the US doing work and visiting friends we ended up in Austin for MWLUG. Another great event organised by Richard Moy and the team with lots of great sessions including Scott Souder’s session on IBM Verse, more on Project Toscana and Ben Langhinrichs’ on Data Visualisation which is an area I’m working a lot in right now.
I had three presentations during the conference and ended up doing four to fill in for a session that was cancelled at the last minute. The Adminblast session I gave was one I hadn’t looked at in over a year until 20 minutes before I started so we all went on a magical journey discovering what I meant to say on each slide as it appeared.
Austin was a great town which I didn’t get to see enough of but luckily we arrived early on the Saturday before the rains started and walked around enjoying the bars and the music. Of all the amazing food on offer I will miss the Vegan Nom taco truck the most. Now to try and reproduce those flavours at home…
IBM Traveler, Management and Security
The SSL Problem and How To Deploy SHA2 (with Mark Myers from LDC Via)
Adminblast Emergency MWLUG Session (original co-authored with Paul Mooney)
Deploying Instant Messaging For Mobile Devices