You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
For [CLAS12 `swif` workflow](https://github.com/baltzell/clas12-workflow) integration, the `bin/run-monitoring.sh` script (which normally generates `slurm` jobs) has a specific mode `--swifjob`:
123
+
For [CLAS12 `swif` workflow](https://github.com/baltzell/clas12-workflow) integration, the `bin/thyme histogram` command (which normally generates `slurm` jobs) has a specific mode `--swifjob`:
124
124
```bash
125
-
bin/run-monitoring.sh --swifjob --focus-detectors # generate files for detector timelines
126
-
bin/run-monitoring.sh --swifjob --focus-physics # generate files for physics QA timelines
125
+
thyme histogram --swifjob --focus-detectors # generate files for detector timelines
Either or both of these commands is _all_ that needs to be executed on a runner node, within [`clas12-workflow`](https://github.com/baltzell/clas12-workflow); calling one of these will automatically run the wrapped code, with the following assumptions and conventions:
129
129
- input HIPO files are at `./` and only a single run will be processed
@@ -149,7 +149,7 @@ top_level_target_directory
149
149
│
150
150
└── ...
151
151
```
152
-
For compatibility with the file tree expected by downstream `bin/run-*-timelines.sh` scripts (see above), symbolic links may be made to these `timeline_{detectors,physics}` directories, but this is not required since these scripts also allow for the specification of an input directory.
152
+
For compatibility with the file tree expected by downstream `thyme analysis` (see above), symbolic links may be made to these `timeline_{detectors,physics}` directories, but this is not required since these scripts also allow for the specification of an input directory.
153
153
154
154
Separate `--focus-detectors` and `--focus-physics` options are preferred, since:
155
155
- offers better organization of the contract data between `swif` jobs and downstream scripts
Copy file name to clipboardExpand all lines: doc/procedure.md
+13-15Lines changed: 13 additions & 15 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -15,17 +15,17 @@ This step reads input HIPO files (_e.g._, DST or `mon` files) and produces histo
15
15
16
16
This step can either be run during the usual data cooking procedure, using [`clas12-workflow`](https://github.com/baltzell/clas12-workflow) (see its usage guide), or it may be run separately on already-cooked data using:
17
17
```bash
18
-
bin/run-monitoring.sh
18
+
thyme histogram
19
19
```
20
20
Running it with no arguments will print the usage guide; use the `--help` option for more detailed guidance.
21
21
22
22
> [!NOTE]
23
-
> If you are performing physics QA for QADB, consider using [**prescaled trains**](/qa-physics/prescaler) (and `run-monitoring.sh` will need the `--flatdir` argument)
23
+
> If you are performing physics QA for QADB, consider using [**prescaled trains**](/qa-physics/prescaler) (and `thyme histogram` will need the `--flatdir` argument)
24
24
25
25
### Example
26
-
If using `clas12-workflow`, see it's documentation; otherwise if using `run-monitoring.sh`:
26
+
If using `clas12-workflow`, see it's documentation; otherwise if using `thyme histogram`:
sbatch ./slurm/job.rga_sp19_v5.physics.slurm # for physics timelines
37
37
```
38
38
- monitor progress with Slurm tools (e.g., `squeue -u $LOGNAME`)
39
-
- monitor output logs in `/farm_out/$LOGNAME/` or use `bin/error-print.sh`
39
+
- monitor output logs in `/farm_out/$LOGNAME/` or use `thyme error`
40
40
41
41
> [!NOTE]
42
42
> - data monitoring for detector timelines is handled by the [`org.jlab.clas.timeline.histograms` package](/src/main/java/org/jlab/clas/timeline/histograms)
@@ -48,8 +48,8 @@ sbatch ./slurm/job.rga_sp19_v5.physics.slurm # for physics timelines
48
48
After Step 1 is complete, run the following Step 2 scripts to produce the timeline HIPO files and to run the automatic QA procedures. There is one script for each timeline type: run them with no arguments to print the usage guides:
49
49
50
50
```bash
51
-
bin/run-detectors-timelines.sh
52
-
bin/run-physics-timelines.sh
51
+
thyme analysis # for detector timelines
52
+
run-physics-timelines.sh# for physics timelines (will eventually be combined with 'thyme analysis')
53
53
```
54
54
55
55
> [!IMPORTANT]
@@ -58,18 +58,16 @@ bin/run-physics-timelines.sh
58
58
> - use the scripts' `-o` option to set the output locations
59
59
60
60
### Example
61
-
**If** you used `clas12-workflow` for Step 1:
61
+
**If** you used `clas12-workflow` for Step 1, the script arguments should be
62
62
```bash
63
-
bin/run-detectors-timelines.sh -d rga_sp19_v5 -i /path/to/output/files # for detector timelines
64
-
bin/run-physics-timelines.sh -d rga_sp19_v5 -i /path/to/output/files # for physics timelines
63
+
-d rga_sp19_v5 -i /path/to/output/files
65
64
```
66
65
- the dataset is _given_ the name `"rga_sp19_v5"` (and does not have to be related to any name given from Step 1)
67
66
- the output from `clas12-workflow` is `/path/to/output/files`; its subdirectories should be run numbers
68
67
69
68
**Otherwise**, you may omit the `-i /path/to/output/files` option (unless you customized it from Step 1):
70
69
```bash
71
-
bin/run-detectors-timelines.sh -d rga_sp19_v5 # for detector timelines
72
-
bin/run-physics-timelines.sh -d rga_sp19_v5 # for physics timelines
70
+
-d rga_sp19_v5
73
71
```
74
72
- the dataset name must match that of Step 1, otherwise you need to specify the path to the input files with `-i`
To view the timelines on the web, you must deploy them by copying the timeline HIPO files to a directory with a running web server. Note that you must have write-permission for that directory. To deploy, run (with no arguments, for the usage guide):
87
85
88
86
```bash
89
-
bin/deploy-timelines.sh
87
+
thyme deploy
90
88
```
91
89
92
90
If all went well, a URL for the new timelines will be printed; open it in a browser to view them.
93
91
94
92
### Example
95
93
```bash
96
-
bin/deploy-timelines.sh -d rga_sp19_v5 -t rga/sp19/pass0/v5 -D # deploy to a run-group web directory (for chefs)
94
+
thyme deploy -d rga_sp19_v5 -t rga/sp19/pass0/v5 -D # deploy to a run-group web directory (for chefs)
97
95
### or ###
98
-
bin/deploy-timelines.sh -d rga_sp19_v5 -t $LOGNAME/my_test -D # deploy to a personal web directory (for testing)
96
+
thyme deploy -d rga_sp19_v5 -t $LOGNAME/my_test -D # deploy to a personal web directory (for testing)
99
97
```
100
98
- this will _only_ print what will be done: deploy the timelines from dataset `"rga_sp19_v5"` (defined in previous step(s)) to the printed path
101
99
- you must have write access to that path; contact the maintainers if you need help with this
Copy file name to clipboardExpand all lines: qa-physics/README.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -43,13 +43,13 @@ For run-group specific notes, including the commands used to perform the QA, see
43
43
* if you find that the DAQ-gated FC charge is larger than the ungated
44
44
charge, you may have assumed here that the recharge option was ON, when
45
45
actually it was OFF and needs to be ON
46
-
*`../bin/run-monitoring.sh`: runs `monitorRead.groovy` on DSTs using `slurm`
46
+
*`thyme histogram`: runs `monitorRead.groovy` on DSTs using `slurm`
47
47
***IMPORTANT**: call this first with the `--check-cache` option to make sure that ALL required DST files are cached; if all files are on `/cache`, you may proceed, removing the `--check-cache` option
48
48
***IMPORTANT**: if you do **NOT** want to analyze full DSTs, and prefer to analyze trains:
49
49
* use the option `--flatdir`, since likely all the HIPO files are in a single directory
50
50
* use the scripts in the [`prescaler/` directory](prescaler) if you want to create a random "prescale" train
51
51
* wait for `slurm` jobs to finish
52
-
* inspect error logs (_e.g._, `../bin/error-print.sh`) to make sure all jobs ran successfully
52
+
* inspect error logs (_e.g._, `thyme error`) to make sure all jobs ran successfully
53
53
*`../bin/run-physics-timelines.sh $dataset`, which does the following:
0 commit comments