https://ldas-jobs.ligo.caltech.edu/~detchar/summary/1164556817-1187733618/range/

seems to have broken links to images or missing plots.

VET is now crashing at CIT, LHO and LLO with "ImportError: cannot import name data_table"

I was able to successfully run at LHO on November 8th, and an identical run now fails.

/home/bhughey/public_html/VET/SN/dec23test/vetcommandline.sh

produces the error below:

------------------------------------------------------------------------------
Welcome to the GW summary information system command-line interface
------------------------------------------------------------------------------

This is process 511252
You have selected gps mode.
Start time 2015-12-23 09:59:43.000 (1134900000.0)
End time: 2015-12-28 01:06:23.000 (1135300000.0)
Output directory: /home/bhughey/public_html/VET/SN/dec23test/1134900000-1135300000

Loading configuration...
Traceback (most recent call last):
  File "/home/detchar/opt/gwpysoft-2.7/bin/gw_summary", line 473, in <module>
    plugins = config.load_plugins()
  File "/home/detchar/opt/gwpysoft-2.7/lib/python2.7/site-packages/gwsumm/config.py", line 241, in load_plugins
    mods.append(import_module(plugin))
  File "/usr/lib64/python2.7/importlib/__init__.py", line 37, in import_module
    __import__(name)
  File "/usr/lib64/python2.7/site-packages/ROOT.py", line 352, in _importhook
    return _orig_ihook( name, *args, **kwds )
  File "/home/detchar/opt/gwpysoft-2.7/lib/python2.7/site-packages/gwvet/tabs.py", line 34, in <module>
    from gwsumm.html import (data_table, markup)
ImportError: cannot import name data_table

The 'Channel Information' section at the bottom of the DataTab should display the frame-type used to read a channel.

None of the coherence spectrograms, or coherence components are being archived. This is trivially because they are stored in a different globalv dict, either we should just store them in the standard globalv.SPECTROGRAMS dict, or add the coherence dicts to gwsumm.archive.

The caption formatting in bootstrap-ligo has been updated (see PR: duncanmmacleod/bootstrap-ligo#1), the submodule can be updated to incorporate this change

Right now the summary pages are ordering tabs in alphabetical order, which isn't great when you want to order based on numbers (i.e. 10 comes before 2). For example see the category 3 tab on:

https://ldas-jobs.ligo-wa.caltech.edu/~laura.nuttall/detchar/O1/VET_hveto_vdf/test_few_hours/1135652417-1135699217/category_3/

It would be useful to display the 'current' GPS time (or the time when the job was started, or similar) on all time-domain plots as a dashed line (probably black), to give a clear display of how up-to-date are the data.

I was trying to make GW summary pages from using GWOSC frames.

https://www.gw-openscience.org/detector_status/1164556817-1187733618/about/index.html

An example ini file that I'm using is:

[calendar]
start-of-week = sunday
start-date = 2015-09-14

[tab-mydata]
name = Test
1 = H1:H1:GWOSC-16KHZ_R1_STRAIN spectrum
2 = L1:L1:GWOSC-16KHZ_R1_STRAIN spectrum

[H1:LOSC-STRAIN]
frametype = H1_LOSC_16_V1

[L1:LOSC-STRAIN]
frametype = L1_LOSC_16_V1

[datafind]
server = 10.5.2.4:80

where that server is what is returned when I use:

$ echo ${LIGO_DATAFIND_SERVER}
10.5.2.4:80

I think I still may be able to run GWsumm through the ligo frame cache files (.lcf), but I'm still working to see if that's possible to do.

Both Marissa and I are unable to run VET at LLO due to what appears to be a matrix transpose error in the plotting code. This includes trying to duplicate runs which worked without error several months ago. The exact same setup runs without generating an error at LHO. Output for run at /home/bhughey/public_html/VET/O1/hveto.a4 is below:

`bash-4.1$ ./vetcommandline.sh
WARNING: AstropyDeprecationWarning: Python 2.6 will no longer be supported from Astropy v1.2.0 and above [astropy]
/mnt/home1/detchar/opt/gwpysoft/lib64/python2.6/site-packages/matplotlib/font_manager.py:273: UserWarning: Matplotlib is building the font cache using fc-list. This may take a moment.
warnings.warn('Matplotlib is building the font cache using fc-list. This may take a moment.')

Welcome to the GW summary information system command-line interface

This is process 2505612
You have selected GPS mode.
Start time 2015-10-20 13:30:00.000 (1129383017.0)
End time: 2015-11-05 10:30:00.000 (1130754617.0)
Output directory: /mnt/home3/bhughey/public_html/VET/O1/hveto.a4/1129383017-1130754617

Loading configuration...
Loaded 1 plugins:
<module 'gwvet.tabs' from '/mnt/home1/detchar/opt/gwpysoft/lib64/python2.6/site-packages/gwvet/tabs.pyc'> Loaded 0 units
Loaded 0 channels
Loaded 7 states
Loaded 12 tabs [2 parents overall]
Reading trigcache from 1 files... done [1 entries]

Processing hveto_pad/Round 6 Pad
Downloaded 72 segments for L1:DMT-ANALYSIS_READY:1 (99.38% coverage).
States finalised [1 total]
Default state: 'Science'
Processing 'Science' state
Read 2451 segments for HVT-RND6P (0.03% coverage).
Veto evaluation results:
Deadtime: 100.0 %
Efficiency: 0.832947709394 %
Efficiency/Deadtime: 0.00832947709394
Efficiency | amplitude>=7: 0.0 %
Efficiency/Deadtime | amplitude>=7: 0.0
Efficiency | amplitude>=8: 0.0 %
Efficiency/Deadtime | amplitude>=8: 0.0
Efficiency | amplitude>=10: 0.0 %
Efficiency/Deadtime | amplitude>=10: 0.0
Use percentage: 0.694160881993 %
Loudest event by snr: 0.0 %
1 data-quality flags identified for segments
Plotting...
/mnt/home1/detchar/opt/gwpysoft/lib64/python2.6/site-packages/matplotlib/patches.py:107: UserWarning: Setting the 'color' property will overridethe edgecolor or facecolor properties.
warnings.warn("Setting the 'color' property will override"
1129383017-1130754617/plots/HV-SCIENCE_4897EE_SEGMENTS-1129383017-1371600.png written
/mnt/home1/detchar/opt/gwpysoft/lib64/python2.6/site-packages/matplotlib/init.py:892: UserWarning: axes.color_cycle is deprecated and replaced with axes.pro
p_cycle; please use the latter.
warnings.warn(self.msg_depr % (key, alt_key))
1129383017-1130754617/plots/L1-SCIENCE_069C73_CWB_TIME_CENTRAL_FREQ_TRIGGERS-1129383017-1371600.png written
/mnt/home1/detchar/opt/gwpysoft/lib64/python2.6/site-packages/matplotlib/axes/_axes.py:5907: UserWarning: 2D hist input should be nsamples x nvariables;
this looks transposed (shape is 1 x 2161)
'(shape is %d x %d)' % inp.shape[::-1])
Traceback (most recent call last):
File "/mnt/home1/detchar/opt/gwpysoft/bin/gw_summary", line 678, in
datafind_error=opts.on_datafind_error, *_cache)
File "/mnt/home1/detchar/opt/gwpysoft/lib64/python2.6/site-packages/gwsumm/tabs/data.py", line 369, in process
*_stateargs)
File "/mnt/home1/detchar/opt/gwpysoft/lib64/python2.6/site-packages/gwvet/tabs.py", line 317, in process_state
return super(FlagTab, self).process_state(state, _args, *_kwargs)
File "/mnt/home1/detchar/opt/gwpysoft/lib64/python2.6/site-packages/gwsumm/tabs/data.py", line 575, in process_state
plot.process()
File "/mnt/home1/detchar/opt/gwpysoft/lib64/python2.6/site-packages/gwsumm/plot/triggers.py", line 473, in process
ax.hist(arr, label=label, weights=weights, *_kwargs)
File "/mnt/home1/detchar/opt/gwpysoft/lib64/python2.6/site-packages/gwpy/plotter/histogram.py", line 109, in hist
return super(HistogramAxes, self).hist(x, *_kwargs)
File "/mnt/home1/detchar/opt/gwpysoft/lib64/python2.6/site-packages/matplotlib/init.py", line 1812, in inner
return func(ax, _args, *_kwargs)
File "/mnt/home1/detchar/opt/gwpysoft/lib64/python2.6/site-packages/matplotlib/axes/_axes.py", line 5970, in hist
raise ValueError('weights should have the same shape as x')
ValueError: weights should have the same shape as x`

Currently the default state is given in the configuration file as the first in the list, however, it might be nicer to allow the user to manually order the states whilst picking a default using a trailing asterisk, as follows:

[tab-hoft]
states = Locked, Ready*, Observing

My student James Pratt and I are currently unable to run VET on CIT, LHO or LLO as of July 8th, using the same setup that worked sometime earlier in the same week.

After sourcing /home/detchar/opt/gwpysoft/etc/nds2-client-user-env.sh here is output of a run at LLO:

/home/detchar/opt/gwpysoft-2.7/lib/python2.7/site-packages/matplotlib/font_manager.py:273: UserWarning: Matplotlib is building the font cache using fc-list. This may take a moment.
warnings.warn('Matplotlib is building the font cache using fc-list. This may take a moment.')

Welcome to the GW summary information system command-line interface

This is process 3007803
You have selected GPS mode.
Start time 2015-10-08 00:00:00.000 (1128297617.0)
End time: 2015-10-21 00:00:00.000 (1129420817.0)
Output directory: /mnt/home3/bhughey/public_html/VET/O1/hveto_test/1128297617-1129420817

Loading configuration...
Loaded 1 plugins:
<module 'gwvet.tabs' from '/home/detchar/opt/gwpysoft-2.7/lib/python2.7/site-packages/gwvet/tabs.pyc'> Loaded 0 units
Loaded 0 channels
Loaded 7 states
Loaded 0 rcParams
Loaded 6 tabs [2 parents overall]
Reading trigcache from 1 files... done [1 entries]

Processing hveto_pad/Round 7: PSL-PMC_MIXER_OUT_DQ Pad
Downloaded 54 segments for L1:DMT-ANALYSIS_READY:1 (99.32% coverage).
States finalised [1 total]
Default state: 'Science'
Processing 'Science' state
Read 8 segments for HVT-RND7P (0.00% coverage).
Traceback (most recent call last):
File "/home/detchar/opt/gwpysoft-2.7/bin/gw_summary", line 682, in
datafind_error=opts.on_datafind_error, *_cache)
File "/home/detchar/opt/gwpysoft-2.7/lib/python2.7/site-packages/gwsumm/tabs/data.py", line 370, in process
*_stateargs)
File "/home/detchar/opt/gwpysoft-2.7/lib/python2.7/site-packages/gwvet/tabs.py", line 305, in process_state
cache=cache, tablename=self.table)
TypeError: get_triggers() got an unexpected keyword argument 'tablename'

The help message for the --ifo argument in gw_summary is ambiguous:

    -i IFO, --ifo IFO   Instrument to process. If this option is set in the
                        [DEFAULT] of any of the INI files, giving it here is
                        redundant. (default: None)

It should ideally mention that you want the prefix for the IFO, and give an example, e.g. L1.

See this example.

Refer to gwpy/gwpy#122.

The summary pages are the specific application of gwpy spectrograms that raised my concern about this issue.

It would be nice if a tab labeled 'Home' could be added to the list of priority tabs so that it could be listed before Summary in the summary page header.

In all current VET runs at CIT or LHO, VET is producing very sparse plots that miss the vast majority of cWB triggers, and typically showing 0% efficiency. (VET at LLO is still broken completely as it apparently hasn't been updated)

As a concrete example,
new/bad page:
https://ldas-jobs.ligo.caltech.edu/~bhughey/VET/cWB/hveto.a4test/1129383017-1130754617/hveto/round_1/

old/good page with identical setup, other than sourcing old shell script in lieu of "activate":
https://ldas-jobs.ligo.caltech.edu/~bhughey/VET/cWB/hveto.a4/1129383017-1130754617/hveto/round_1/

(In other current runs, VET may show more than one trigger, but the plots are always very sparse)

It would be great to update the HvetoTab to display the results from the new python-hveto, perhaps with a prototype on GEO data.

I tested this summary page following the Wiki's instruction page. I just type the command:
gw_summary --nds --verbose --ifo L1 --output-dir /home/${USER}/public_html/detchar --verbose --config-file ${RUN_DIR}/defaults.ini --config-file ${RUN_DUR}/ihope.ini --max-processes 40 --day 20140222
usage: gw_summary [-h] [-V] {week,gps,day,month} ...
gw_summary: error: argument mode: invalid choice: 'L1' (choose from 'week', 'gps', 'day', 'month')

and get an error message, showing L1 ifo option. But I don't see what's going on here.

Adding the --nds causes a gw_summary error (I'm using your version of gw_summary, from /home/detchar/...):

[max.isi@ldas-pcdev3 ~]$ gw_summary day 20150115 --ifo C1 --config-file /home/max.isi/summary/config/defaults.ini -f /home/max.isi/summary/config/c1hoft.ini --multi-process 10 --nds --verbose

------------------------------------------------------------------------------
Welcome to the GW summary information system command-line interface
------------------------------------------------------------------------------

You have selected DAY mode.
Start time 2015-01-15 00:00:00.000 (1105315216.0)
End time: 2015-01-16 00:00:00.000 (1105401616.0)
Output directory: /mnt/qfs4/max.isi/day/20150115

-------------------------------------------------
Processing Seismic
States finalised
Processing 'All' state
    6 channels identified for TimeSeries
    Fetching data (from nds) for 6 channels [raw]Traceback (most recent call last):
  File "/mnt/qfs5/detchar/opt/gwpysoft/bin/gw_summary", line 731, in <module>
    segdb_error=opts.on_segdb_error)
  File "/mnt/qfs5/detchar/opt/gwpysoft/lib64/python2.6/site-packages/gwsumm/tabs/data.py", line 381, in process
    plotqueue=queue, **stateargs)
  File "/mnt/qfs5/detchar/opt/gwpysoft/lib64/python2.6/site-packages/gwsumm/tabs/data.py", line 447, in process_state
    return_=False)
  File "/mnt/qfs5/detchar/opt/gwpysoft/lib64/python2.6/site-packages/gwsumm/data.py", line 360, in get_timeseries_dict
    **ioargs)
  File "/mnt/qfs5/detchar/opt/gwpysoft/lib64/python2.6/site-packages/gwsumm/data.py", line 519, in _get_timeseries_dict
    **ioargs)
  File "/mnt/qfs5/detchar/opt/gwpysoft/lib64/python2.6/site-packages/gwpy/utils/deps.py", line 81, in wrapper
    return func(*args, **kwargs)
TypeError: fetch() got an unexpected keyword argument 'dtype'

Also, I thought adding an [nds] section in the config file was enough to make it use NDS, but it won't work unless I explicitly add the --nds flag as well...
Thanks,
Max

per @duncanmmacleod's suggestion:

"I had an idea to write the HDF5 archive file after every tab, basically to act as a checkpointing file in case one of the later tabs causes a failure."

When using the gwvet plugin with gw_summary, some trigger scatter plots are missing the IFO prefix in the name, see for example this page which includes this png name:

.../-AFTER_CAT_2_800A18_OMICRON_TIME_PEAK_FREQUENCY_TRIGGERS-1135652417-46800.png

The Segment Information table on a DataTab prints the known and active segment percentages relative to the total defined duration of the tab (end-start). For long-duration pages (like the LIGO O2 pages) the end time is typically a long time in the future, so the percentages are very confusing.

It makes more sense (or at least would be less confusing) to show the percentage relative to time-to-date (now-start) or relative to the known segments, or the state segments, or something like that.

If a segment database is not explicitly given in the config file, gwsumm throws this error:

File "/home/detchar/opt/gwpysoft-2.7/lib/python2.7/site-packages/gwsumm/segments.py", line 154, in get_segments
kwargs['url'] = config.get('segment-database', 'url')
File "/usr/lib64/python2.7/ConfigParser.py", line 607, in get
raise NoSectionError(section)
ConfigParser.NoSectionError: No section: 'segment-database'

Could specifying [segment-database] be made optional?

The H1 Summary Page (https://ldas-jobs.ligo-wa.caltech.edu/~detchar/summary/) either doesn't come up or shows plots with a timestamp from 22+hrs ago.

L1's Summary Page appears to be fine.

Is this the way we report issues with this page? Or is there an email we should send to someone?

Thanks,

c.

PS This interface did not like my @ligo-wa.caltech.edu email address. It worked when I used my ligo.org email.

Currently the 'Report an issue' link in the <footer> for all pages points here (//github.com/gwpy/gwsumm/issues/). It would be good to allow the configuration to override this with a custom URL, defaulting to the current.

Converting from ASD (/rtHz) into amplitude spectrum units requires knowledge of the equivalent noise bandwidth (ENB{W}). It makes sense to add the FFTlength, overlap, window, and ENBW to spectral figures, or at least make it easier to discover from the HTML. ENBW can be calculated from the window and sampling rate as:

def enbw(window, sampling):
    a = window.sum()
    b = (window ** 2).sum()
    return sampling * b / a**2

A simple example throws a ValueError when given the following INI file

[tab-spectra]
name = spectra
1 = L1:IMC-F_OUT_DQ spectrum
1-ylabel = 'Frequency noise [Hz]'

and the following command line

gw_summary gps -f etc/example.ini '00:00' '00:10'

this gives:

... snip ...
  File "/home/duncan.macleod/.local/lib/python2.6/site-packages/gwsumm-0.0.0.dev780-py2.6.egg/gwsumm/data.py", line 820, in _get_spectrogram
    **fftparams)
  File "/mnt/home2/duncan.macleod/.local/lib/python2.6/site-packages/gwpy-0.1b1.dev119-py2.6.egg/gwpy/timeseries/core.py", line 535, in spectrogram
    raise ValueError("overlap must be less than fftlength")
ValueError: overlap must be less than fftlength

From this pages the first link is missing:
https://ldas-jobs.ligo.caltech.edu/~duncan.macleod/gwsumm/latest/tabs/websites.html#websites

More precisely: “As we have seen” which points to: https://ldas-jobs.ligo.caltech.edu/~duncan.macleod/gwsumm/latest/tabs/index

There isn't currently a way to have a summary page tab that acts as a direct link to a URL or a location on disk. This would be immediately useful in being able to create a "home" tab that returns the user to the splash page for the summary pages.

The glue.segments module is also being deprecated in favor of ligo.segments.

It would be nice to have the ability to add captions to figures, so when they are enlarged in a fancybox, there's a bit of explanatory text.

This should be pretty easy for the PlotTab and DataTab given the current INI syntax:

[tab-my-data]
...
1 = H1:GDS-CALIB_STRAIN spectrum
1-caption = 'Amplitude Spectral Density of the calibrated strain readout of the LHO detector'

Thanks to @tjmassin for the inspiration.

Is there any way, or is the feature planned, to be able to look at timescales greater than one day? I.e. week, month, year, custom date ranges?

Personally, I think this could quite useful.

Livingston commissioners and project managers reckon it would be useful to have a way to keep track of locklosses. The specific concern is whether overnight lockloss is worth operators working more or less hours, so having a way to quickly access histograms of lockloss vs. time of day would be useful.

Currently the footer of summary pages says

This page was created by detchar at 09:11 on November 09 2018.

Having the time zone would be very useful. Maybe showing the time with the time zone of the user viewing the page would be even better, if possible!

If, when using multiprocessing, a frame-reading process raises an Exception, these aren't passed back to the top level, and the child process just hangs there.

The correct behaviour is for the top-level to correctly raise the given exception.

This may turn out to be a bug in GWpy.

It would be a nice and useful feature to be able to add thresholds (horizontal and vertical) to figures when no EPICS channels exist for this purpose. Currently the advice is to use 'hline' and 'vline' however there is no support to specify the colour, linestyle and label to them, for instance in the way of; 'hline-linestyle', 'hline-labels', 'hline-color'

There are various configuration options for bootstrap-datepicker for displaying the calendar on summary pages. It might be good to expose some of them to the user in configuration files for gw_summary. An couple of examples would be to expose:

• daysOfWeekHighlighted to allow the user to highlight particular days of the week;
• or, datesDisabled to allow certain dates, for which no page exists, to be disabled.

An application of allowing certain dates to be highlighted, and others disabled, is in #259, which allows a list of given dates to be highlighted, with all the rest disables by default. This makes use of the beforeShowDay option to turn highlighting on for a given selection of dates.

On the BLRMS plots for the seismometers, it seems to me that we should be plotting the mean minute trend rather than the RMS. The RMS operation has already been done in the front end RMS code. Isn't what we want to know the 'root-MEAN-square', rather than the RMS of the RMS?

When using the --nds option, if data are missing gw_summary will return a RuntimeError: Requested data were not found error and crash rather than padding with zeros or producing an empty plot.

cc @duncanmmacleod

The summary pages for LIGO Hanford are configured to have a tab showing coherence spectra between various LSC channels and DARM. This tab is currently failing with the following traceback:

Traceback (most recent call last):
  File "/home/detchar/opt/summary-2.7/bin/gw_summary", line 682, in <module>
    datafind_error=opts.on_datafind_error, **cache)
  File "/home/detchar/opt/summary-2.7/lib/python2.7/site-packages/gwsumm/tabs/data.py", line 366, in process
    **stateargs)
  File "/home/detchar/opt/summary-2.7/lib/python2.7/site-packages/gwsumm/tabs/data.py", line 504, in process_state
    datafind_error=datafind_error, **fp2)
  File "/home/detchar/opt/summary-2.7/lib/python2.7/site-packages/gwsumm/data/utils.py", line 68, in decorated_func
    return f(arg1, segments, *args, **kwargs)
  File "/home/detchar/opt/summary-2.7/lib/python2.7/site-packages/gwsumm/data/coherence.py", line 437, in get_coherence_spectrograms
    datafind_error=datafind_error, **fftparams)
  File "/home/detchar/opt/summary-2.7/lib/python2.7/site-packages/gwsumm/data/utils.py", line 68, in decorated_func
    return f(arg1, segments, *args, **kwargs)
  File "/home/detchar/opt/summary-2.7/lib/python2.7/site-packages/gwsumm/data/coherence.py", line 67, in get_coherence_spectrogram
    **fftparams)
  File "/home/detchar/opt/summary-2.7/lib/python2.7/site-packages/gwsumm/data/utils.py", line 68, in decorated_func
    return f(arg1, segments, *args, **kwargs)
  File "/home/detchar/opt/summary-2.7/lib/python2.7/site-packages/gwsumm/data/coherence.py", line 231, in _get_coherence_spectrogram
    ts2, stride, nproc=nproc, **fftparams)
  File "/home/detchar/opt/summary-2.7/lib/python2.7/site-packages/gwpy/timeseries/timeseries.py", line 722, in csd_spectrogram
    nproc=nproc, **kwargs)
  File "/home/detchar/opt/summary-2.7/lib/python2.7/site-packages/gwpy/signal/fft/ui.py", line 340, in average_spectrogram
    psds = mp_utils.multiprocess_with_queues(nproc, _psd, tschunks)
  File "/home/detchar/opt/summary-2.7/lib/python2.7/site-packages/gwpy/utils/mp.py", line 171, in multiprocess_with_queues
    raise res
ValueError: window is longer than input signal

This error is got with the following command line call:

gw_summary day --on-segdb-error warn --verbose --output-dir . --on-datafind-error warn --multi-process 8 --no-html --ifo H1 --archive H1LSC --config-file /home/detchar/public_html/summary/etc/defaults.ini,/home/detchar/public_html/summary/etc/global.ini,/home/detchar/public_html/summary/etc/lsc.ini,/home/detchar/public_html/summary/etc/h1lsc.ini

Accessing the L1 page requires a password, which is not required for the H1 page.
Is this intentional?

Hi Alex,
I like the plot showing the combined VT in the summary page. But I found a point that makes me wonder why. Please correct me if I am wrong.

The plot about VT in O2 (see https://ldas-jobs.ligo.caltech.edu/~detchar/summary/1164556817-1187733618/plots/H1L1V1-MULTI_D220F1_TIME_VOLUME-1164556817-23176801.png) shows that the Virgo VT is higher than the LIGO VT. Also, the LIGO VTs are zero all the way.

I tested this summary page following the Wiki's instruction page. I just type the command:

gw_summary --nds --verbose --ifo L1 --output-dir /home/${USER}/public_html/detchar --verbose --config-file ${RUN_DIR}/defaults.ini --config-file ${RUN_DUR}/ihope.ini --max-processes 40 --day 20140222
usage: gw_summary [-h] [-V] {week,gps,day,month} ...
gw_summary: error: argument mode: invalid choice: 'L1' (choose from 'week', 'gps', 'day', 'month')

and get an error message, showing L1 ifo option. But I don't see what's going on here.

Whilst at LHO I took at look at some of the isc_lock guardian numbers and noticed they're a bit different to what is in the ini file. I've copied some relevant info below so this can be update. Thanks!

no 13
15 locking als
50 acquire_prmi
51 prmi locked
102 drmi_locked_prep_asc
201 start_tr_carm
202 carm_to_tr
203 carm_150_picometers
No 205
304 darm_to_rf
305 DHARD_WFS
306 park_als_vco
307 shutter_als
308 carm_offset_reduction
No 405,6,8
409 carm_to_refl
425 drmi_to_pop
428 engage rf violins
430 engage_asc_for_full_ifo
no 453
503 engage dc violins
504 reduce rf45 modulation depth
505 increase power
507 adjust power
520 low noise length control
550 close beam diverters
600 nominal low noise

Aidan Brooks suggested we display a cumulative event rate plot on the network all-observing-run page configuration. This will require some software development to do the following:

• Query GraceDb for gravitational-wave event candidates
• Display the histogram of number of events as a function of inverse false alarm rate
• Show three traces: one for EM-bright (or equivalent), one for non-EM-bright CBCs, and a third for all events

The following verbose output is confusing/wrong:

...
    Finding L-L1_HOFT_C00 frames for [1173225610, 1173273256)... 11 found.
    Gaps discovered in aggregated h(t) type L1_HOFT_C00, checking L1_DMT_C00
    Finding L-L1_DMT_C00 frames for [1173225610, 1173273256)... 0 found.
    No extra coverage with frametype L1_DMT_C00
    Fetching data (from frames) for 1 channels [L1_DMT_C00]
...

i.e. the frametype in the final line should still be L1_HOFT_C00 since no extra frames were found for L1_DMT_C00.

Older incarnations of the summary pages supported using a fixed reference spectrum for normalisation of a spectrogram - this should be easily implemented by allowing the ratio parameter for a given figure to point to a file on disk.

On the L1 summary page home, 'isntrument' is spelled wrong. ;) Fix it before someone important notices!

Cheers,

• Dan

Hi Duncan,

at the LVC meeting in Glasgow I had hinted at a small date problem with the summary page.

I think what happens is that the page uses the user timezone to define the link of the "today" button, in the second cascade menu (the calendar).

So if someone is e.g. in Europe and try to click on "today" at 1AM local, it will get something like

Not Found

The requested URL /~detchar/summary/day/20161117/ was not found on this server.

simply because in the US it's still the day before.

Of course, the problem becomes more and more evident as you go East in the globe.

A simple fix could be to always use LHO local time to define what "today" means.

Cheers,
salvo

In order to run properly, gwpy-0.12.0 or greater is required. This should be explicitly required by requirements.txt and setup.py.