Here's an interesting phenomenon that perhaps some of you have observed, or maybe not. I've observed it when running in SRB mode and specifying LINKAGE=BRANCH when using WTOs for messaging. Perhaps someone can explain what is going on here.
For situtation #1, my app is running in SRB mode, it physically resides in ECSA, supervistor state, key 0. I needed to do some debugging and added some strategically placed WTOs. It's a fast running process and many times the WTOs would never externalize. I finally figured out that if coded SYNCH=YES on the WTO, I would see the WTOs in the job output. However, issuing the WTO with SYNCH=YES caused the app to run doggggy slow, and I mean REALLY slooooooow. Situation #2 is similar. I'm testing z/XDC 1.7 SRB mode support, and I coded up a simple SRB, run it in batch, forced it to abend. I then got control with z/XDC and starting stepping through my retry code. I traced around a WTO LINKAGE=BRANCH, jumped over to my other session to browse the output (SDSF "DA"), but the WTO had not shown up. I stepped through code for a couple more minutes, exited the debugging session and then cancelled the job. I did not see the WTO in the job output until I cancelled the job. So my question is, why does it take so long for the WTO to externalize? Thanks, Todd (By the way, the z/XDC testing is going great. I can't remember how many years I wanted a debugger for SRB mode... Well, maybe I can... 15. Picture yourself instruction stepping through code, in supervisor state, key 0, SRB mode, in an EUT=YES FRR, full cross memory mode. It's amazing. Thanks Dave!) ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
