jmayzurk

FWIW, I have the weatherbug module enabled also. -Jeff

Like sk8er02 I'm still experiencing some problems around networking that are definitely new to the later releases. I've been able to replicate twice that simply having the ethernet link go down will cause the ISY99 system date to jump back to the epoch. And then it starts logging continuous NTP errors even after the link is restored. I'm trying to isolate this problem and give you a set of conditions to reproduce. I've got a terminal server on the serial port now also...

Yes, will do. Thanks for the help. -JEff

Michel, Thanks, as usual, for the prompt response. 1. NTP synch is 60 minutes (1 hour in the configuration tab) 2. Static address for the ISY. 3. Gigabit switch which is connected to the router. Other programs work, but the ISY stopped logging after the network stack crashed. -Jeff

I had a similar problem a few days after upgrading to 2.7.15. Same symptoms as sk8er02: No network access to the ISY at all (including telnet). A power cycle brought it back. One thing I noticed is a HUGE number of -60001 errors in the log: Over 16,000 entries in the 36 hours prior to the crash. 0 wolfgan Fri 2010/05/07 12:07:13 AM System -60001 0 wolfgan Fri 2010/05/07 12:07:22 AM System -60001 0 wolfgan Fri 2010/05/07 12:07:31 AM System -60001 0 wolfgan Fri 2010/05/07 12:07:40 AM System -60001 0 wolfgan Fri 2010/05/07 12:07:49 AM System -60001 [jeffm@optimator Desktop]$ fgrep 60001 log-20100505.txt | wc -l 16192 The ISY had internet access during this time, and in fact, I was using MobiLinc Pro during this period. My ISY locked up once before with an earlier release, but without all the -60001 errors in the log. -Jeff

With my v2 thermostat adapter + Venstar T1800: - If fan is manually turned on (button on T1800), ISY99i reflects the state change to "On" - If fan is turned on automatically because the system is heating or cooling, fan state does not change; it remains in "auto" Aside: Perhaps the control is inappropriately labeled? It seems this is more of a "fan mode" control rather than "fan state" (i.e., is the fan running or not?). -Jeff

Michel, I verified that I can upload using the latest JWS admin console app on my Mac (MacOS 10.6.3, Java 1.6.0_17 64-bit) against my ISY99i running 2.7.14RC1. I re-loaded 2.7.14RC1 on the first try. Thanks for the quick fix! -Jeff

Michel, I know there have been a ton of posts on this topic already, but methinks you guys must have broken something around 2.7.12. The last two releases have been a real pain to upload. Neither one would upload from my Mac browsers (Safari 4.0.5 and Firefox 3.6.2), so I resorted to IE7 on a PC. When upgrading to 2.7.14, it failed twice using IE7 also--but eventually worked. I tried both HTTP and HTTPS, and even increased the timeouts for both using the telnet console. Nothing made a difference. -Jeff

Michael, The 64-bit version is listed first in both sections. My java version: [jeffm@fuji Desktop]$ java -version java version "1.6.0_17" Java(TM) SE Runtime Environment (build 1.6.0_17-b04-248-10M3025) Java HotSpot(TM) 64-Bit Server VM (build 14.3-b01-101, mixed mode) [/code]

I have the same problem, since about 2.7.10a I believe. It happens on multiple machines over both SSL and HTTP ports. I typically launch the JWS app directly rather than going through a browser. MacOS 10.6.2, no firewall software. -Jeff

FYI, I upgraded my ISY99i from 2.7.9 to 2.7.10 and everything seems to be working fine, including v2 thermostat adapter which was added during 2.7.9. I did not need to restore from a pre-2.7.9 backup. So far so good! -Jeff

I just ran into this myself, and it seems strange. It seems that the "if" block is re-evaluted during execution of the "then" block. I have a program like this: IF (backdoor light is off AND garage door sensor is on) THEN { turn on backdoor light turn on some other lights wait 15 minutes turn off backdoor light turn off some other lights } But the everything after the wait statement in the THEN block never gets executed, presumably because a condition in the IF block is no longer true. Is this by design, or is it perhaps a bug? Maybe it's just the old C programmer in me, but it seems like control should remain with the THEN block until it has completed executing. Otherwise there is the possibility of an asynchronous event interrupting my program in the middle of execution and leaving my devices in the wrong state (e.g., projector on but screen still up). In other words, it allows for race conditions. -Jeff