I heard of IOC other than ROC. There was a misconception for me in this early morning. I apologize.
I know how to deal with this situation.
situation 1: 8:27 When we tried to start run 3633, start run failed. We kill coda and retried. (That's correct. Luis told me you can type "kcoda" three times, if it shows that you have to type "Y" or "N". Just type "Y", and repeat "kcoda" three times to make sure of clearing the coda system is killed. Then, start typing "startcoda".) after clicking on "configuration", wait for a moment to see if all of ROC scripts are running and connected completely. Then, click on "download". Still, please pay attention to make sure of all of ROCs running. No rush. Somehow, one or two ROCs are not responding and just stuck there. How can we do? see situation 2. (situation 1: go to PREx How-To page: "http://hallaweb.jlab.org/parity/prex/runinfo/opmanual.html and find" --> Rebooting HAPPEX/PREX crates
situation 2: start run 3634 failed: roc31 not reporting. Rebooted roc31. Before typing "reboot" for one given ROC, please turn the parity coda off. close the coda window. Then, reboot the coda. Wait for a while until it shows you "WORKING", and then type "kcoda" again before "startcoda". Still, after clicking "configuration", make sure of the script of every of ROCs are working. Then, click on "download", make sure the ROC that you just rebooted are running without the error message.
situation 3: somehow, it still doesn't work after typing "reboot" for one given ROC. Don't worry! Just go to PREx "How-To" web-page http://hallaweb.jlab.org/parity/prex/runinfo/opmanual.html --> Power-Cycling the Happex/PREX Crates After power cycling one or two ROCs (ROC31 usually died with ROC24 or ROC25), just follow "Rebooting HAPPEX/PREX crates" procedure. Please do remember: Don't turn the parity coda on before making sure of the rebooted ROC working.
Crate Location Outlet Crate/Portserver Injector 1 g0inj Injector 2 happex
situation 4: ROC 23 issue ROC23 is another issue about "dithering". Somehow, it doesn't work in the middle of data-taking. Don't stop the run. Try turn the dithering off, then type "reboot". If it still doesn't work, and other ROCs work quite well, let this production run finish, as we still have good parity data. We just cannot see dithering plot on prompt. After this run finish, turn the CODA off. "reboot" or "power-cycle" ROC23. Make sure of its working well, then "kcoda" --> "startcoda" --> "configuration" (look at ROC23 script connected --> "download" (look at ROC23 running.... and the downloading is successfully.) --> "start" (look at ROC23 "pre-start..." and then "go....")
situation 5: the data flow is jumping like a triangle please stop this run. turn the coda off. make sure of all of ROCs are still alive without any error message. Then, "kcoda"-->"startcoda"-->"configuraton" (see if all of ROCs running smoothly)-->"download" (then, probably the problematic ROC will show up.) Don't go ahead. "cacel" this run. Reboot this ROC and then "kcoda" again before typing "startcoda".
I am sorry I said "I don't know", as I thought I really have no idea of "IOC" problem.