| | 1 | Notes I took while trying to get the simtest running on IPP 2.1 using a Linux machine. It took a really long time to get the whole thing running successfully, but eventually it did finish. Below are notes I jotted down about errors or configuration changes. - Julia Fang |
| | 2 | |
| | 3 | == Configuration == |
| | 4 | |
| | 5 | * Pre-simtest preparations |
| | 6 | Before proceeding to the simtest, I first created a mysql database, configured and set up ippMonitor on my webpage (ippMonitor software found under ~/ipp-2.1/ippMonitor), and downloaded the full grizy astrometry database from the folks over at Manoa. |
| | 7 | |
| | 8 | * The .ipprc file |
| | 9 | I went to my ippconfig folder (located ~/psconfig/default.linrh/share/ippconfig/) and copied "ipprc.config" to my home directory and rename it as ".ipprc" |
| | 10 | |
| | 11 | * .ptolemyrc file |
| | 12 | I copied dvo.site from my ippconfig folder to my home directory and renamed it .ptolemyrc. |
| | 13 | |
| | 14 | * A good way to check for configuration files |
| | 15 | How to see if there are any more configuration files that need to be modified (general or specific to a camera): |
| | 16 | > ppConfigDump -camera SIMTEST -dump-recipe PSASTRO - |
| | 17 | You can change PSASTRO to whatever you're looking for, i.e. REJECTIONS. |
| | 18 | |
| | 19 | == Data Injection == |
| | 20 | |
| | 21 | * IPC/CMD.pm is not in the right directory |
| | 22 | My reference star database (full grizy from Manoa, Hawaii) path needed to be specified under my ippconfig folder in 2 locations: |
| | 23 | ippconfig/recipes/psastro.config |
| | 24 | ippconfig/simtest/psastro.config |
| | 25 | I went to those files and edited their DVO.CATDIR line so that they pointed to where my astrometry database is located. |
| | 26 | |
| | 27 | == Pantasks Setup == |
| | 28 | |
| | 29 | * Pantasks can't find config file: some functions will be unavailable. |
| | 30 | Gene said this was optional and not necessary, so I didn't worry about it. |
| | 31 | |
| | 32 | * How to setup my cluster |
| | 33 | I went to ipp-2.1/ippTasks/pantasks.pro and added my cluster information. I looked at the other examples there for help. Here's mine: |
| | 34 | |
| | 35 | macro init.cluster.hl |
| | 36 | $parallel = 1 |
| | 37 | controller exit true |
| | 38 | controller host add koloa |
| | 39 | controller host add shadowfax |
| | 40 | end |
| | 41 | |
| | 42 | I also had to make sure all the computers had access to my full grizy catalog as well as permissions to my mysql database. |
| | 43 | |
| | 44 | == Detrend Construction == |
| | 45 | |
| | 46 | * In the event that I need to redo everything and start over |
| | 47 | To blow away all of my tables and start on this database all over again: |
| | 48 | > pxadmin -recreate -dbname nameofdb |
| | 49 | |
| | 50 | * How to turn on/off specific tasks in pantasks |
| | 51 | |
| | 52 | detproc.on, detproc.off |
| | 53 | detresid.on, detresid.off |
| | 54 | detstack.on, detstack.off |
| | 55 | detnorm.on, detnorm.off |
| | 56 | |
| | 57 | And so forth. I can type in "status' to see what's on and what's off. Also looked at ippMonitor to see which steps have completed and which were still processing. |
| | 58 | |
| | 59 | * dettool.stack.run would never finish |
| | 60 | I quit pantasks to clear the buffers, then reloaded pantasks, modules, and clusters. To see the jobs that get launched, I typed in: |
| | 61 | > $VERBOSE = 3 |
| | 62 | I loaded all the tasks but only left the detrend step on. Set pantasks to run and looked carefully for the "detrend_stack.pl" command and tried running it outside of pantasks in a new terminal. |
| | 63 | |
| | 64 | * Issues with dettool.resid.load and reject tool while trying to get master dark |
| | 65 | I started over again by re-runing the whole detrend process and eventually worked through all the tools (either via pantasks or by hand). |
| | 66 | |
| | 67 | == Chip and Camera == |
| | 68 | |
| | 69 | * Reference stars wouldn't load in pantasks |
| | 70 | I had to run each image outside of pantasks so the reference stars could be loaded for the chip step. |
| | 71 | |
| | 72 | * test0066 and test0073 had a problem: "apSum is negative" |
| | 73 | This was not actually an error, just merely informative. I turned it off by going to my ~/.ipprc file and commented out: |
| | 74 | # err S32 10 |
| | 75 | and changed the loglevel from 9 to 4. |
| | 76 | |
| | 77 | * Camera step didn't work |
| | 78 | Solved by downloading current version of Ohana from Paul Price, and copying the new version of "getstar" to /default.linrh/bin/. |
| | 79 | |
| | 80 | == Warp == |
| | 81 | |
| | 82 | * Where do I put my skycells? |
| | 83 | Anywhere is fine, as long as I specify it in SIMTEST's specific camera.config folder (/share/ippconfig/simtest/camera.config) |
| | 84 | |
| | 85 | * Warp stage not completing in pantasks |
| | 86 | I made sure that my grizy catalog directory was specified in psastro.config (both /recipes and /simtest). |
| | 87 | |
| | 88 | * Error in warp_skycell.pl, ppStats error |
| | 89 | Fixed by going to .ipprc and making sure that the "simple" camera is always uncommented and last in the list as a good catch-all camera. |
| | 90 | |
| | 91 | == Difference and Stack == |
| | 92 | |
| | 93 | No issues. |