Indico v2.2.8 LaTex errors

On our new indico v2.2.8 installation we see latex errors like the following (the output.log (36.5 KB) file has been attached).
Latex was installed using the indico instructions and I have added explicitly the STRICT_LATEX = False in the indico.conf.

I would appreciate your help.

Subject: Unexpected Exception occurred at Could not compile '/opt/indico/tmp/indico-texgen-MdFa_g/single_doc.tex'. Read '/opt/indico/tmp/indico-texgen-MdFa_g/output.log' for details

2020-05-10 06:06:18,202  c04affde9e50435d  indico.flask - ERROR -- Could not compile '/opt/indico/tmp/indico-texgen-MdFa_g/single_doc.tex'. Read '/opt/indico/tmp/indico-texgen-MdFa_g/output.log' for details

Traceback (most recent call last):
  File "/opt/indico/.venv/lib/python2.7/site-packages/flask/", line 1949, in full_dispatch_request
    rv = self.dispatch_request()
  File "/opt/indico/.venv/lib/python2.7/site-packages/flask/", line 1935, in dispatch_request
    return self.view_functions[rule.endpoint](**req.view_args)
  File "/opt/indico/.venv/lib/python2.7/site-packages/indico/web/flask/", line 84, in wrapper
    return obj().process()
  File "/opt/indico/.venv/lib/python2.7/site-packages/indico/web/", line 275, in process
    res = self._do_process()
  File "/opt/indico/.venv/lib/python2.7/site-packages/indico/modules/events/controllers/", line 54, in _do_process
    return RHEventBase._do_process(self)
  File "/opt/indico/.venv/lib/python2.7/site-packages/indico/web/", line 245, in _do_process
    rv = self._process()
  File "/opt/indico/.venv/lib/python2.7/site-packages/indico/modules/events/contributions/controllers/", line 184, in _process
    return send_file('contribution.pdf', pdf.generate(), 'application/pdf')
  File "/opt/indico/.venv/lib/python2.7/site-packages/indico/legacy/pdfinterface/", line 62, in generate
    return, **self._args)
  File "/opt/indico/.venv/lib/python2.7/site-packages/indico/legacy/pdfinterface/", line 199, in run
    raise LaTeXRuntimeException(source_filename, log_filename)
LaTeXRuntimeException: LaTeX compilation of /opt/indico/tmp/indico-texgen-MdFa_g/single_doc.tex failed

{u'data': {u'get': {},
           u'headers': {'Accept': u'text/html,application/xhtml+xml,application/signed-exchange;v=b3,application/xml;q=0.9,*/*;q=0.8',
                        'Accept-Encoding': u'gzip,deflate,br',
                        'Amp-Cache-Transform': u'google;v="1..3"',
                        'Connection': u'keep-alive',
                        'From': u'googlebot(at)',
                        'Host': u'',
                        'User-Agent': u'Mozilla/5.0 (Linux; Android 6.0.1; Nexus 5X Build/MMB29P) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2272.96 Mobile Safari/537.36 (compatible; Googlebot/2.1; +'},
           u'json': None,
           u'post': {},
           u'url': {'confId': u'13745', 'contrib_id': 21484}},
 u'endpoint': u'contributions.export_pdf',
 u'id': 'c04affde9e50435d',
 u'ip': '',
 u'method': 'GET',
 u'referrer': None,
 u'rh': 'RHContributionExportToPDF',
 u'time': '2020-05-10T06:06:18.226830',
 u'url': u'',
 u'user': None,
 u'user_agent': u'Mozilla/5.0 (Linux; Android 6.0.1; Nexus 5X Build/MMB29P) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2272.96 Mobile Safari/537.36 (compatible; Googlebot/2.1; +'}


This seems wrong, in filenames the _ should not get escaped. As this repeats, it looks there is a problem in the temp name generation for the conference logo.

And I actually do not see why the _ is there at all.

Can you post the exact source of the contribution’s description?

I assume that what you are requesting is in the error log. This event is public, so the contribution is:

and the indico entry is:

What I meant is the text you get when you edit the contribution and copy the contents of the “description” field. Basically the raw source code of the contribution where nothing has been rendered to html or latex.

Actually, works fine for me now. If something was changed in there in the meantime that makes figuring out what the actual problem was much harder now…

@ThiefMaster What you want is what is in the database?

Yes, but if you click on edit and copy it from there it’s much easier to get than by going to the DB…

This is the text:

Based on the premise that the neutrino beam used in DUNE traverses the Earth’s crust on its journey from being produced until detection, we propose a non-standard interaction (NSI) between the neutrinos and the matter of the Earth’s crust. Such NSI can cause the change of the flavor and change the energy of the neutrinos. The phenomenon of the flavor oscillations that we know can produce changes that may be measurable from the analysis of the oscillation parameters. In this work we study how the proposed NSI can change the sensitivity of the DUNE in measuring the values of the parameters δCP, θ13, and θ23. NSI also affect the determination of the hierarchy of the neutrino masses. 
For this study we performed simulations using the General Long Baseline Experiment Simulator (GLoBES). We determined optimal values for the oscillation parameters from the literature and tested some values for δCP. We used a method of χ² to test different values of δCP, θ13, and θ23. This way we were able to determine regions of sensitivity. We chose nine distinct forms of mass density distribution in the baseline and determined how the oscillation parameters change for each of them. When we compared the various sensitivity curves we elaborated a relative χ² for the purpose of determining the influence of the distribution of matter. Finally, we determined which values the parameters describing the NSI εαβ should be in order to be similar to the interactions we know.   
With these studies, we conclude that the matter density distribution does not significantly change the sensitivity of the DUNE in measuring the parameters δCP, θ13, and θ23, however, the presence of the NSI causes sufficiently large changes in the sensitivity of the DUNE. NSI is also affected by mass hierarchy and, in this way, it can help to determine it. After an accurate study of NSI on the propagation, we intend to study NSI in the interaction, and how it may affect the events in LArTPCs.