SmartestHouse Posted November 14, 2015 Posted November 14, 2015 I've upgraded from an ISY-99 to an 994 and also replaced the modem at the same time. (Turned out the modem was broken, but given the age of the ISY I chose to replace that at the same time.) I restored from a backup. There is a remaining problem: The ISY doesn't seem to notice when a device gets manually switched on or off. The device's status doesn't update in the ISY and programs don't get triggered by devices. If I manually refresh the device status or switch it from the ISY it works fine. So it seems the ISY can talk to my devices, but it doesn't listen to their messages about a switch being pressed and so on. Any ideas how to fix this?
larryllix Posted November 14, 2015 Posted November 14, 2015 (edited) Have you checked each device's links and compared them with ISYs links for accuracy? This is available by right-clicking on each device in the device tree in the Admin Console. Edited November 14, 2015 by larryllix
SmartestHouse Posted November 14, 2015 Author Posted November 14, 2015 Have you checked each device's links and compared them with ISYs links for accuracy? It seems there's one link in each device that's not the same, and restoring the device doesn't change that. For example, I have: Device: <isy.diag><title>Device Links Table : Office / 1A 17 13 1</title><insteon.lincs><eventInfo> <ix>0</ix> <ad>4088</ad> <fl>162</fl> <gr>0</gr> <id>3905241</id> <data>16719617</data> </eventInfo> <eventInfo> <ix>1</ix> <ad>4080</ad> <fl>162</fl> <gr>19</gr> <id>3905241</id> <data>16719617</data> </eventInfo> <eventInfo> <ix>2</ix> <ad>4072</ad> <fl>0</fl> <gr>0</gr> <id>1834330</id> <data>16719617</data> </eventInfo> </insteon.lincs></isy.diag> ISY: <isy.diag><title>ISY Links Table : Office / 1A 17 13 1</title><insteon.lincs><eventInfo> <ix>0</ix> <ad>4088</ad> <fl>162</fl> <gr>0</gr> <id>3905241</id> <data>16719617</data> </eventInfo> <eventInfo> <ix>1</ix> <ad>4080</ad> <fl>162</fl> <gr>19</gr> <id>3905241</id> <data>16719617</data> </eventInfo> <eventInfo> <ix>2</ix> <ad>4072</ad> <fl>0</fl> <gr>0</gr> <id>0</id> <data>0</data> </eventInfo> </insteon.lincs></isy.diag>
LeeG Posted November 14, 2015 Posted November 14, 2015 The last link record is logically the same, End of List, Office / 1A 17 13 1 0FF8 : A2 00 3B.96.D9 FF 1F 01 0FF0 : A2 13 3B.96.D9 FF 1F 01 0FE8 : 00 00 1B.FD.5A FF 1F 01 Office / 1A 17 13 1 0FF8 : A2 00 3B.96.D9 FF 1F 01 0FF0 : A2 13 3B.96.D9 FF 1F 01 0FE8 : 00 00 00.00.00 00 00 00 This device has no Controller link record(s) (start with E2) so the ISY will not receive device change messages. It looks like the PLM was Deleted somewhere along the process. The corrective action is to File | Restore ISY from a backup taken before PLM was Deleted, followed by a File | Restore Devices.
Techman Posted November 14, 2015 Posted November 14, 2015 It seems there's one link in each device that's not the same, and restoring the device doesn't change that. Do a "Show Device Links Table" then click on compare. The results will show you any mismatched links.
SmartestHouse Posted November 14, 2015 Author Posted November 14, 2015 This device has no Controller link record(s) (start with E2) so the ISY will not receive device change messages. It looks like the PLM was Deleted somewhere along the process. The corrective action is to File | Restore ISY from a backup taken before PLM was Deleted, followed by a File | Restore Devices. Thanks! That was what I did in the first place to set up the new ISY, but somehow doing the same thing again resolved it.
larryllix Posted November 14, 2015 Posted November 14, 2015 (edited) The last link record is logically the same, End of List, Office / 1A 17 13 1 0FF8 : A2 00 3B.96.D9 FF 1F 01 0FF0 : A2 13 3B.96.D9 FF 1F 01 0FE8 : 00 00 1B.FD.5A FF 1F 01 Office / 1A 17 13 1 0FF8 : A2 00 3B.96.D9 FF 1F 01 0FF0 : A2 13 3B.96.D9 FF 1F 01 0FE8 : 00 00 00.00.00 00 00 00 This device has no Controller link record(s) (start with E2) so the ISY will not receive device change messages. It looks like the PLM was Deleted somewhere along the process. The corrective action is to File | Restore ISY from a backup taken before PLM was Deleted, followed by a File | Restore Devices. Ouch! Thanks for jumping in there LeeG! I confess I have never seen that before. Where does that code chart come from? Edited November 14, 2015 by larryllix
LeeG Posted November 14, 2015 Posted November 14, 2015 (edited) larryllix As far as Insteon devices are concerned only the first byte of a link record being 0x00 is necessary to indicate End of List. With many Insteon devices now in use still using Peek/Poke to update link records (1 byte at a time) it would be a waste to update the other 7 bytes for appearance only. I don't think the "ISY" data actually takes up space for 8 bytes of zeros. The ISY knows how many link records are active (2 in this case) and displays 8 bytes of zeros after the last active link record. See the attachment Compare display which indicates the End of List as [ignore]. The link record at 0FA8 in attachment looks like a mismatch but is Insteon using two bits in first byte for Hops management. Just like the End of List, other link records can appear bad but be fine because Insteon I2CS devices use bits in the first byte that are both temporary and not told to the PLM/ISY. Office / 1A 17 13 1 - from device 0FF8 : A2 00 3B.96.D9 FF 1F 01 0FF0 : A2 13 3B.96.D9 FF 1F 01 0FE8 : 00 00 1B.FD.5A FF 1F 01 Office / 1A 17 13 1 - from ISY data 0FF8 : A2 00 3B.96.D9 FF 1F 01 0FF0 : A2 13 3B.96.D9 FF 1F 01 0FE8 : 00 00 00.00.00 00 00 00 EDIT: sorry, don't know what "code chart" is referring to. Edited November 14, 2015 by LeeG
jerlands Posted November 14, 2015 Posted November 14, 2015 The last link record is logically the same, End of List, Office / 1A 17 13 1 0FF8 : A2 00 3B.96.D9 FF 1F 01 0FF0 : A2 13 3B.96.D9 FF 1F 01 0FE8 : 00 00 1B.FD.5A FF 1F 01 Office / 1A 17 13 1 0FF8 : A2 00 3B.96.D9 FF 1F 01 0FF0 : A2 13 3B.96.D9 FF 1F 01 0FE8 : 00 00 00.00.00 00 00 00 This device has no Controller link record(s) (start with E2) so the ISY will not receive device change messages. It looks like the PLM was Deleted somewhere along the process. The corrective action is to File | Restore ISY from a backup taken before PLM was Deleted, followed by a File | Restore Devices. How do you transpose from <xml> code (or whatever it is) back into the link record... e.g., <eventInfo> <ix>1</ix> <ad>4080</ad> <fl>162</fl> <gr>19</gr> <id>3905241</id> <data>16719617</data> = 0FF0 : A2 13 3B.96.D9 FF 1F 01 Jon...
larryllix Posted November 14, 2015 Posted November 14, 2015 larryllix As far as Insteon devices are concerned only the first byte of a link record being 0x00 is necessary to indicate End of List. With many Insteon devices now in use still using Peek/Poke to update link records (1 byte at a time) it would be a waste to update the other 7 bytes for appearance only. I don't think the "ISY" data actually takes up space for 8 bytes of zeros. The ISY knows how many link records are active (2 in this case) and displays 8 bytes of zeros after the last active link record. See the attachment Compare display which indicates the End of List as [ignore]. The link record at 0FA8 in attachment looks like a mismatch but is Insteon using two bits in first byte for Hops management. Just like the End of List, other link records can appear bad but be fine because Insteon I2CS devices use bits in the first byte that are both temporary and not told to the PLM/ISY. ... EDIT: sorry, don't know what "code chart" is referring to. Oh shoot! I should not have included YOUR text in my question. Sorry for the confusion. I was referring to what looks like XML code as per Jerlands above also.
LeeG Posted November 14, 2015 Posted November 14, 2015 jerlands, larryllix The XML to link record display is a very quick, very dirty Java program i wrote months ago because I was tired of converting a very long posted XML file. The ISY Saves the link file display in XML which is fine, had to be in some format that could be written. However, for the human trying to analyze one of the large XML files, it was a pain.
jerlands Posted November 15, 2015 Posted November 15, 2015 jerlands, larryllix The XML to link record display is a very quick, very dirty Java program i wrote months ago because I was tired of converting a very long posted XML file. The ISY Saves the link file display in XML which is fine, had to be in some format that could be written. However, for the human trying to analyze one of the large XML files, it was a pain. That's interesting... but ah, err.. how does one go about analyzing with the human tool any resources available? Thanks, jon...
larryllix Posted November 15, 2015 Posted November 15, 2015 jerlands, larryllix The XML to link record display is a very quick, very dirty Java program i wrote months ago because I was tired of converting a very long posted XML file. The ISY Saves the link file display in XML which is fine, had to be in some format that could be written. However, for the human trying to analyze one of the large XML files, it was a pain. Thanks Lee, but where does the XML come from? Is it from an unzipped backup file?
LeeG Posted November 15, 2015 Posted November 15, 2015 The "Show Device Links Table" and "Show ISY Links Table" have a Save button on the displayed link table. The Save function writes XML file.
jerlands Posted November 15, 2015 Posted November 15, 2015 The "Show Device Links Table" and "Show ISY Links Table" have a Save button on the displayed link table. The Save function writes XML file. I wonder what the reason is behind this when it's so difficult for most to do anything with? Jon...
LeeG Posted November 15, 2015 Posted November 15, 2015 I suspect most who understand Insteon and link tables debug from the display popup and post the displayed image.
stusviews Posted November 15, 2015 Posted November 15, 2015 I wonder what the reason is behind this when it's so difficult for most to do anything with? Jon... Interpreting code to text uses resources that are better used providing features.
Recommended Posts