Forum Replies Created
-
AuthorReplies
-
5.8.4 on AIX 6.1
We handle 200KB+ uuencoded Muse content without problem.
A TPS Inbound Data proc extracts the uuencoded content from the ZPD segment, regsubs the “\X0D\\X0A\” strings to “n”, converts the HL7 escape sequences (that’s actually a little tricky) and then execs uudecode.
I can send you the code if you’re interested.
5.8.4 on AIX 6.1
We handle 200KB+ uuencoded Muse content without problem.
A TPS Inbound Data proc extracts the uuencoded content from the ZPD segment, regsubs the “\X0D\\X0A\” strings to “n”, converts the HL7 escape sequences (that’s actually a little tricky) and then execs uudecode.
I can send you the code if you’re interested.
Apologies for any ambiguities in the use of the term “cluster”. I was referring to high availability, two-node, active/passive clusters. If anyone has implemented cloverleaf on a (to me) more complex load-balancing cluster (with high availability), I would be interested to know more about your configuration.
Chas,
If you are on a unix platform you should be able to run tcpdump and observe exactly watch is happening on the socket.
Sometimes a “bad” transaction will cause the receiving application to shutdown/disconnect the socket. This is very easy to observe/prove with the output of tcpdump. If the receiving application sends a tcp socket disconnect packet/message back to cloverleaf immediately after receiving the transaction that Cloverleaf sent to it, that sequence of events should be apparent in the output of tcpdump. You can then send a screen scrape to the vendor and ask why their application is doing that.
–Bob
Our Advanced Security Server runs on a Windows 2003 server so it goes down once a month for patching. There is no disruption to running interfaces but there is no IDE activity allowed during the outages.
We are not using https. Outside access to Global Monitor, if ever required, would go through a user’s VPN connection and use a remote desktop app. Remote developers and oncall folks haven’t had a need for GM.
We are on AIX 6.1.
The data center folks have not complained about any timeouts. And they know my number. 😀
We’ve had Global Monitor in place with 5.8.4 for a couple of months. Primarily got it to help the data center folks keep an eye on lots of new sites. Another benefit is that it allows a group of non-Cloverleafers to investigate, edit and replay SMAT transactions. Other groups can check out and stop/start the interfaces to their non-production applications. A very small group is setup to check a few production interfaces after monthly patching of their applications.
It’s a web-based app so there is no client installation – that’s nice. I wish its security worked with Cloverleaf’s advanced security. Instead, a user’s certificate and key need to be stored on the Global Monitor server.
Jim,
I’d be curious to see what tcpdump shows.
I am experiencing the same problem when the Security Server (a Win 2008 VM) reboots. It appears there is a newer hcihostid.exe that is part of the 5.8.4 (and maybe earlier) patch. The newer exe returns the same host ID after a reboot.
Thanks Jim.
Thanks Jim. Short and sweet!
😀 I should have caught the missing bracket. I over-focused on the : 😯
Russ,
Thanks for all the info.
AIx’s HACMP design is different than VMware’s HA.
I’m still curious about how the hcihostid is derived on AIX.
–Bob
Russ,
I don’t think there’s a host serial number in the Intel world.
Code:uname -a
Linux clovertestl.mayo.edu 2.6.9-89.0.9.ELsmp #1 SMP Wed Aug 19 08:07:15 EDT 2009 i686 i686 i386 GNU/LinuxI’d be curious what
Code:lscfg -vl ent0
shows on your AIX host. -
AuthorReplies