Quantcast

fc-cache fails after recent MikTeX update

classic Classic list List threaded Threaded
11 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

fc-cache fails after recent MikTeX update

Jan Přikryl
Dear all,

since the last "major" MiKTeX update approximately two weeks ago, MiKTeX
fails to completely install a new package (be it on-the-fly or through
MiKTeX package manager). On manual install it always complains about

MiKTeX Problem Report
Message: The executed process did not succeed.
Data: fileName="C:\TeX\MiKTeX 2.9\miktex\bin\initexmf.exe",
arguments="--admin --mklinks --mkmaps", exitCode="1"
Source: Libraries\MiKTeX\Core\Process\Process.cpp
Line: 159
MiKTeX: 2.9
OS: Windows 7 Enterprise, 64-bit, Service Pack 1, build 7601
Invokers: .../explorer/PSPad/explorer
SystemAdmin: yes
PowerUser: no
Root0: C:\Users\Jan Přikryl\AppData\Roaming\MiKTeX\2.9
Root1: C:\Users\Jan Přikryl\AppData\Local\MiKTeX\2.9
Root2: C:\ProgramData\MiKTeX\2.9
Root3: C:\TeX\MiKTeX 2.9
UserInstall: C:\Users\Jan Přikryl\AppData\Roaming\MiKTeX\2.9
UserConfig: C:\Users\Jan Přikryl\AppData\Roaming\MiKTeX\2.9
UserData: C:\Users\Jan Přikryl\AppData\Local\MiKTeX\2.9
CommonInstall: C:\TeX\MiKTeX 2.9
CommonConfig: C:\ProgramData\MiKTeX\2.9
CommonData: C:\ProgramData\MiKTeX\2.9

The logs unfortunately do not provide much information. I have only
information from initexmf, which reads as follows:

2016-05-31 09:01:20,333+0200 INFO  initexmf - starting: MiKTeX
Configuration Utility 2.9.5900 (MiKTeX 2.9)
2016-05-31 09:01:20,503+0200 INFO  initexmf - running: mkfntmap --admin
2016-05-31 09:01:33,201+0200 FATAL initexmf - The executed process did
not succeed.
2016-05-31 09:01:33,201+0200 FATAL initexmf - Info:
fileName="C:\TeX\MiKTeX 2.9\miktex\bin\mkfntmap.EXE",
arguments="--admin", exitCode="1"
2016-05-31 09:01:33,202+0200 FATAL initexmf - Source:
Libraries\MiKTeX\Core\Process\Process.cpp
2016-05-31 09:01:33,202+0200 FATAL initexmf - Line: 159

This is consistent with the error report from the installer, but as far
as I understoot the lines listed above, the initextmf error is that a
subprocess "mkfntmap" returned with non-zero exit code and that's it.
The mkfntmap.log error contains the following:

2016-05-31 09:01:20,859+0200 INFO  mkfntmap - Entering administrative
mode...
2016-05-31 09:01:20,873+0200 INFO  mkfntmap - Parsing config file
"C:\TeX\MiKTeX 2.9\miktex\config\updmap.cfg"...
2016-05-31 09:01:20,959+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\tetex\dvips35.map"...
2016-05-31 09:01:20,968+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\tetex\pdftex35.map"...
2016-05-31 09:01:20,977+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\tetex\dvipdfm35.map"...
2016-05-31 09:01:20,986+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\tetex\ps2pk35.map"...
2016-05-31 09:01:21,264+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\cm-super\cm-super-t1.map"...
2016-05-31 09:01:21,275+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\cm-super\cm-super-t2a.map"...
2016-05-31 09:01:21,287+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\cm-super\cm-super-t2b.map"...
2016-05-31 09:01:21,298+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\cm-super\cm-super-t2c.map"...
2016-05-31 09:01:21,310+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\cm-super\cm-super-ts1.map"...
2016-05-31 09:01:21,321+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\cm-super\cm-super-x2.map"...
2016-05-31 09:01:21,332+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\amsfonts\cm.map"...
2016-05-31 09:01:21,340+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\amsfonts\cmextra.map"...
2016-05-31 09:01:21,425+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\amsfonts\cyrillic.map"...
2016-05-31 09:01:21,469+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\dvips\dstroke\dstroke.map"...
2016-05-31 09:01:21,557+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\amsfonts\euler.map"...
2016-05-31 09:01:21,570+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\dvips\eurosym\eurosym.map"...
2016-05-31 09:01:21,705+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\amsfonts\latxfont.map"...
2016-05-31 09:01:21,829+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\mflogo-font\mflogo.map"...
2016-05-31 09:01:21,998+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\rsfs\rsfs.map"...
2016-05-31 09:01:22,115+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\amsfonts\symbols.map"...
2016-05-31 09:01:22,295+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\xypic\xypic.map"...
2016-05-31 09:01:22,830+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\adforn\OrnementsADF.map"...
2016-05-31 09:01:23,500+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\arabi\arabi.map"...
2016-05-31 09:01:23,854+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\psnfss\charter.map"...
2016-05-31 09:01:24,441+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\psnfss\fpls.map"...
2016-05-31 09:01:25,055+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\lm\lm.map"...
2016-05-31 09:01:25,155+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\obsolete\mathpi.map"...
2016-05-31 09:01:25,163+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\obsolete\mathpple-ext.map"...
2016-05-31 09:01:25,801+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\psnfss\pazo.map"...
2016-05-31 09:01:25,829+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\dvips\pslatex\pcrr8rn.map"...
2016-05-31 09:01:25,946+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\psnfss\psnfss.map"...
2016-05-31 09:01:26,425+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\avantgar\uag.map"...
2016-05-31 09:01:26,459+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\bookman\ubk.map"...
2016-05-31 09:01:26,477+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\courier\ucr.map"...
2016-05-31 09:01:26,516+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\helvetic\uhv.map"...
2016-05-31 09:01:26,540+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\ncntrsbk\unc.map"...
2016-05-31 09:01:26,565+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\palatino\upl.map"...
2016-05-31 09:01:26,605+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\times\utm.map"...
2016-05-31 09:01:26,613+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\psnfss\utopia.map"...
2016-05-31 09:01:26,622+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\zapfchan\uzc.map"...
2016-05-31 09:01:26,631+0200 INFO  mkfntmap - Parsing "C:\TeX\MiKTeX
2.9\fonts\map\dvips\zapfding\uzd.map"...
2016-05-31 09:01:26,936+0200 INFO  mkfntmap - Writing
C:\ProgramData\MiKTeX\2.9\dvips\config\ps2pk.map...
2016-05-31 09:01:26,967+0200 INFO  mkfntmap - Writing
C:\ProgramData\MiKTeX\2.9\dvips\config\download35.map...
2016-05-31 09:01:26,969+0200 INFO  mkfntmap - Writing
C:\ProgramData\MiKTeX\2.9\dvips\config\builtin35.map...
2016-05-31 09:01:26,970+0200 INFO  mkfntmap - Writing
C:\ProgramData\MiKTeX\2.9\dvips\config\psfonts_t1.map...
2016-05-31 09:01:26,990+0200 INFO  mkfntmap - Writing
C:\ProgramData\MiKTeX\2.9\dvips\config\psfonts_pk.map...
2016-05-31 09:01:27,003+0200 INFO  mkfntmap - Writing
C:\ProgramData\MiKTeX\2.9\pdftex\config\pdftex_ndl14.map...
2016-05-31 09:01:27,022+0200 INFO  mkfntmap - Writing
C:\ProgramData\MiKTeX\2.9\pdftex\config\pdftex_dl14.map...
2016-05-31 09:01:27,041+0200 INFO  mkfntmap - Writing
C:\ProgramData\MiKTeX\2.9\dvipdfm\config\dvipdfm_dl14.map...
2016-05-31 09:01:27,056+0200 INFO  mkfntmap - Writing
C:\ProgramData\MiKTeX\2.9\dvipdfm\config\dvipdfm_ndl14.map...
2016-05-31 09:01:27,071+0200 INFO  mkfntmap - Copying
C:\ProgramData\MiKTeX\2.9\dvips\config\psfonts_t1.map
2016-05-31 09:01:27,071+0200 INFO  mkfntmap -      to
C:\ProgramData\MiKTeX\2.9\dvips\config\psfonts.map...
2016-05-31 09:01:27,073+0200 INFO  mkfntmap - Copying
C:\ProgramData\MiKTeX\2.9\dvipdfm\config\dvipdfm_dl14.map
2016-05-31 09:01:27,073+0200 INFO  mkfntmap -      to
C:\ProgramData\MiKTeX\2.9\dvipdfm\config\dvipdfm.map...
2016-05-31 09:01:27,074+0200 INFO  mkfntmap - Copying
C:\ProgramData\MiKTeX\2.9\pdftex\config\pdftex_dl14.map
2016-05-31 09:01:27,074+0200 INFO  mkfntmap -      to
C:\ProgramData\MiKTeX\2.9\pdftex\config\pdftex.map...
2016-05-31 09:01:27,112+0200 INFO  mkfntmap - running: fc-cache
--miktex-admin
2016-05-31 09:01:33,172+0200 INFO  mkfntmap - C:/Program Files
(x86)/Adobe/Acrobat Reader DC/Resource/Font/PFM: failed to write cache
2016-05-31 09:01:33,172+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1: failed to write cache
2016-05-31 09:01:33,172+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1/adobe: failed to write cache
2016-05-31 09:01:33,172+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1/arabi: failed to write cache
2016-05-31 09:01:33,172+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1/arkandis: failed to write cache
2016-05-31 09:01:33,172+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1/arkandis/adforn: failed to write cache
2016-05-31 09:01:33,172+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1/public: failed to write cache
2016-05-31 09:01:33,173+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1/public/amsfonts: failed to write cache
2016-05-31 09:01:33,173+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1/public/cm-super: failed to write cache
2016-05-31 09:01:33,173+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1/urw: failed to write cache
2016-05-31 09:01:33,173+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/opentype: failed to write cache
2016-05-31 09:01:33,173+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/opentype/public: failed to write cache
2016-05-31 09:01:33,173+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1: error scanning
2016-05-31 09:01:33,173+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1/adobe: error scanning
2016-05-31 09:01:33,173+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1/arabi: error scanning
2016-05-31 09:01:33,173+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1/arkandis: error scanning
2016-05-31 09:01:33,173+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1/public: error scanning
2016-05-31 09:01:33,174+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1/public/amsfonts: error scanning
2016-05-31 09:01:33,174+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/type1/urw: error scanningJ
2016-05-31 09:01:33,174+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/opentype: error scanning
2016-05-31 09:01:33,176+0200 INFO  mkfntmap - C:/TeX/MiKTeX
2.9/fonts/opentype/public: error scanning

... and that's it. At the end it seems that the culprit is the
"fc-cache" tool which probably chokes on the fact that it shall create
font information caches in directories that do not contain any fonts. It
seems to me as the fc-cache option --error-on-no-fonts would be
implicitly given. The fc-cache.log is of course empty ...

Any clue what I can do, short of reinstalling MiKTeX?

Thanks.

--
  Jan Přikryl
  [hidden email]

------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are
consuming the most bandwidth. Provides multi-vendor support for NetFlow,
J-Flow, sFlow and other flows. Make informed decisions using capacity
planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
_______________________________________________
Q: How can I leave the mailing list?
A: See http://docs.miktex.org/faq/support.html#leavingml
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: fc-cache fails after recent MikTeX update

Ulrike Fischer-2
Am Tue, 31 May 2016 09:20:02 +0200 schrieb Jan Přikryl:

> since the last "major" MiKTeX update approximately two weeks ago, MiKTeX
> fails to completely install a new package (be it on-the-fly or through
> MiKTeX package manager). On manual install it always complains about

> UserInstall: C:\Users\Jan Přikryl\AppData\Roaming\MiKTeX\2.9

I saw a similar error message on tex.sx and had a discussion about
it in the chat, but nothing definitive was found.

He also made a bug report but the error couldn't be reproduced:

https://sourceforge.net/p/miktex/bugs/2468/

Like you this user had a non-ascii chars in the user name, so it is
possible that this leads to the failure of fc-cache to write the
cache.



--
Ulrike Fischer
http://www.troubleshooting-tex.de/


------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are
consuming the most bandwidth. Provides multi-vendor support for NetFlow,
J-Flow, sFlow and other flows. Make informed decisions using capacity
planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
_______________________________________________
Q: How can I leave the mailing list?
A: See http://docs.miktex.org/faq/support.html#leavingml
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: fc-cache fails after recent MikTeX update

Juris Freimanis
Dear all,

I seem the latest big MiKTeX update (launched in the middle of May 2016)
is a complete failure. E-mails with messages about serious errors come
from many MiKTeX users. For myself: 1) the compilation does not work on
my computer (neither .tex -> .dvi nor .tex -> .pdf), 2) MiKTeX
administrative update (copystart_admin.exe) ends with the error message:

=============================

Sorry, but something went wrong:

Too many open files: path = "C:\Program Files\MiKTeX
2.9\tpm\packages\miktex-metapost-bin-2.9.tpm"

path = "C:\Program Files\MiKTeX
2.9\tpm\packages\miktex-metapost-bin-2.9.tpm"

=============================

I have Windows 7 Professional SP1 as my operating system, with the
latest Windows updates installed. It is clear that it will be necessary
to deinstall and completely reinstall MiKTeX if I still wish to use
MiKTeX in future. However, while there are so many error messages from
other colleagues I am not sure if it is worth to spend my time for
reinstalling MiKTeX.

I am not a MiKTeX developer; I am completely uninterested to study the
complicated directory structure of MiKTeX software, the role of each
.dll library etc. My only wish is to have some usable tool in order to
produce scientific papers in my branch. I ask Christian Schenk and all
the other relevant people to restore on all servers the MiKTeX version
which was operative until the beginning of May 2016. Please launch only
such MiKTeX updates which have been thoroughly tested.

Best wishes,

Juris Freimanis

On 2016.05.31. 12:08, Ulrike Fischer wrote:

> Am Tue, 31 May 2016 09:20:02 +0200 schrieb Jan Přikryl:
>
>> since the last "major" MiKTeX update approximately two weeks ago, MiKTeX
>> fails to completely install a new package (be it on-the-fly or through
>> MiKTeX package manager). On manual install it always complains about
>> UserInstall: C:\Users\Jan Přikryl\AppData\Roaming\MiKTeX\2.9
> I saw a similar error message on tex.sx and had a discussion about
> it in the chat, but nothing definitive was found.
>
> He also made a bug report but the error couldn't be reproduced:
>
> https://sourceforge.net/p/miktex/bugs/2468/
>
> Like you this user had a non-ascii chars in the user name, so it is
> possible that this leads to the failure of fc-cache to write the
> cache.
>
>
>


------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are
consuming the most bandwidth. Provides multi-vendor support for NetFlow,
J-Flow, sFlow and other flows. Make informed decisions using capacity
planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
_______________________________________________
Q: How can I leave the mailing list?
A: See http://docs.miktex.org/faq/support.html#leavingml
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: fc-cache fails after recent MikTeX update

Jan Přikryl-2
In reply to this post by Ulrike Fischer-2
Dne Út, 31. květen 2016 v 11:08 h uživatel Ulrike Fischer napsal:

> I saw a similar error message on tex.sx and had a discussion about
> it in the chat, but nothing definitive was found.

Aaah, I think I remember that message, especially it's first version ...

> He also made a bug report but the error couldn't be reproduced:
>
> https://sourceforge.net/p/miktex/bugs/2468/
>
> Like you this user had a non-ascii chars in the user name, so it is
> possible that this leads to the failure of fc-cache to write the
> cache.

Very well. Can I get a hold of a debug version of fc-cache to confirm or
disprove that? I have tried to produce some meaningful  report by
setting MIKTEX_TRACE to "files", starting Sysinternals DbgView and
running

"C:\TeX\MiKTeX 2.9\miktex\bin\fc-cache.exe" --verbose --verbose
--verbose --miktex-admin

but the information received is zero (the debug view is completely
empty).

I have also tried to look at the files that fc-cache.exe tries to read
using Sysinternals Process Monitor, but I am unable to find anything
suspicious.

--
  Jan Přikryl
  [hidden email]

------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are
consuming the most bandwidth. Provides multi-vendor support for NetFlow,
J-Flow, sFlow and other flows. Make informed decisions using capacity
planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
_______________________________________________
Q: How can I leave the mailing list?
A: See http://docs.miktex.org/faq/support.html#leavingml
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: fc-cache fails after recent MikTeX update

Ulrike Fischer-2
Am Tue, 31 May 2016 12:14:29 +0200 schrieb Jan Přikryl:

> Dne Út, 31. květen 2016 v 11:08 h uživatel Ulrike Fischer napsal:
>
>> I saw a similar error message on tex.sx and had a discussion about
>> it in the chat, but nothing definitive was found.
>
> Aaah, I think I remember that message, especially it's first version ...
>
>> He also made a bug report but the error couldn't be reproduced:
>>
>> https://sourceforge.net/p/miktex/bugs/2468/
>>
>> Like you this user had a non-ascii chars in the user name, so it is
>> possible that this leads to the failure of fc-cache to write the
>> cache.
>
> Very well. Can I get a hold of a debug version of fc-cache to confirm or
> disprove that? I have tried to produce some meaningful  report by
> setting MIKTEX_TRACE to "files", starting Sysinternals DbgView and
> running
>
> "C:\TeX\MiKTeX 2.9\miktex\bin\fc-cache.exe" --verbose --verbose
> --verbose --miktex-admin
>
> but the information received is zero (the debug view is completely
> empty).

You need to start debug view with admin rights to get output (the
tests can then be done in some cmd window with user rights).
 
>
> I have also tried to look at the files that fc-cache.exe tries to read
> using Sysinternals Process Monitor, but I am unable to find anything
> suspicious.

It looks more like a writing problem "failed to write cache".

The process monitor log of the other user didn't show any suspicious
too, but I'm not sure if he showed everything.

> Any clue what I can do, short of reinstalling MiKTeX?

If you reinstall miktex you should setup the user roots  so that
they don't contain non-ascii chars in the name. But it would be
naturally better to get at the source of the problem. Add a comment
to the bug report that you are affected too.


--
Ulrike Fischer
http://www.troubleshooting-tex.de/


------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are
consuming the most bandwidth. Provides multi-vendor support for NetFlow,
J-Flow, sFlow and other flows. Make informed decisions using capacity
planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
_______________________________________________
Q: How can I leave the mailing list?
A: See http://docs.miktex.org/faq/support.html#leavingml
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Too many open files (was fc-cache fails after recent MikTeX update)

Jan Přikryl-2
In reply to this post by Juris Freimanis
Juris,

first, be aware that you hijacked my post ;-).

Second, monitor the bug report 2499, it probably describes your problem
as well.

Third, like it or not, MiKTeX is free of charge, and developed by a
single developer,  so you have to live with occasional periods of issues
until things settle down. Even it these periods are more frequent in the
last half a year (at least for me). I am not happy with that as well,
but I have no solution short of a) paying a full-time developer to help
Christian (but a decent developer in CZ would cost you some 50-60
thousand EUR per year, which is approximately fivefold of my salary as a
university Assistant Professor) in stabilising the current code base or
b) do it myself (for which I do not have time)...

-- jan

Dne Út, 31. květen 2016 v 11:47 h uživatel Juris Freimanis napsal:

> Dear all,
>
> I seem the latest big MiKTeX update (launched in the middle of May 2016)
> is a complete failure. E-mails with messages about serious errors come
> from many MiKTeX users. For myself: 1) the compilation does not work on
> my computer (neither .tex -> .dvi nor .tex -> .pdf), 2) MiKTeX
> administrative update (copystart_admin.exe) ends with the error message:
>
> =============================
>
> Sorry, but something went wrong:
>
> Too many open files: path = "C:\Program Files\MiKTeX
> 2.9\tpm\packages\miktex-metapost-bin-2.9.tpm"
>
> path = "C:\Program Files\MiKTeX
> 2.9\tpm\packages\miktex-metapost-bin-2.9.tpm"
>
> =============================
>
> I have Windows 7 Professional SP1 as my operating system, with the
> latest Windows updates installed. It is clear that it will be necessary
> to deinstall and completely reinstall MiKTeX if I still wish to use
> MiKTeX in future. However, while there are so many error messages from
> other colleagues I am not sure if it is worth to spend my time for
> reinstalling MiKTeX.
>
> I am not a MiKTeX developer; I am completely uninterested to study the
> complicated directory structure of MiKTeX software, the role of each
> .dll library etc. My only wish is to have some usable tool in order to
> produce scientific papers in my branch. I ask Christian Schenk and all
> the other relevant people to restore on all servers the MiKTeX version
> which was operative until the beginning of May 2016. Please launch only
> such MiKTeX updates which have been thoroughly tested.
>
> Best wishes,
>
> Juris Freimanis
>
> On 2016.05.31. 12:08, Ulrike Fischer wrote:
> > Am Tue, 31 May 2016 09:20:02 +0200 schrieb Jan Přikryl:
> >
> >> since the last "major" MiKTeX update approximately two weeks ago, MiKTeX
> >> fails to completely install a new package (be it on-the-fly or through
> >> MiKTeX package manager). On manual install it always complains about
> >> UserInstall: C:\Users\Jan Přikryl\AppData\Roaming\MiKTeX\2.9
> > I saw a similar error message on tex.sx and had a discussion about
> > it in the chat, but nothing definitive was found.
> >
> > He also made a bug report but the error couldn't be reproduced:
> >
> > https://sourceforge.net/p/miktex/bugs/2468/
> >
> > Like you this user had a non-ascii chars in the user name, so it is
> > possible that this leads to the failure of fc-cache to write the
> > cache.
> >
> >
> >
>
>
> ------------------------------------------------------------------------------
> What NetFlow Analyzer can do for you? Monitors network bandwidth and
> traffic
> patterns at an interface-level. Reveals which users, apps, and protocols
> are
> consuming the most bandwidth. Provides multi-vendor support for NetFlow,
> J-Flow, sFlow and other flows. Make informed decisions using capacity
> planning reports.
> https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
> _______________________________________________
> Q: How can I leave the mailing list?
> A: See http://docs.miktex.org/faq/support.html#leavingml


--
  Jan Přikryl
  [hidden email]

------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are
consuming the most bandwidth. Provides multi-vendor support for NetFlow,
J-Flow, sFlow and other flows. Make informed decisions using capacity
planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
_______________________________________________
Q: How can I leave the mailing list?
A: See http://docs.miktex.org/faq/support.html#leavingml
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Too many open files (was fc-cache fails after recent MikTeX update)

Juris Freimanis
Jan,

I have not hijacked anything. I received e-mail from Ulrike Fischer
today at 12:08 East European Daylight Saving Time (= GMT + 3h). Ulrike's
e-mail was addressed to the whole list
[hidden email], including myself. Your e-mail was
automatically cited In that Ulrike's e-mail. I wrote my e-mail as a
reply to Ulrike's e-mail, and my mailing program (Mozilla Thunderbird)
automatically cited the previous e-mails (both Ulrike's and yours).

Juris Freimanis


On 2016.05.31. 13:53, Jan Přikryl wrote:

> Juris,
>
> first, be aware that you hijacked my post ;-).
>
> Second, monitor the bug report 2499, it probably describes your problem
> as well.
>
> Third, like it or not, MiKTeX is free of charge, and developed by a
> single developer,  so you have to live with occasional periods of issues
> until things settle down. Even it these periods are more frequent in the
> last half a year (at least for me). I am not happy with that as well,
> but I have no solution short of a) paying a full-time developer to help
> Christian (but a decent developer in CZ would cost you some 50-60
> thousand EUR per year, which is approximately fivefold of my salary as a
> university Assistant Professor) in stabilising the current code base or
> b) do it myself (for which I do not have time)...
>
> -- jan
>
> Dne Út, 31. květen 2016 v 11:47 h uživatel Juris Freimanis napsal:
>> Dear all,
>>
>> I seem the latest big MiKTeX update (launched in the middle of May 2016)
>> is a complete failure. E-mails with messages about serious errors come
>> from many MiKTeX users. For myself: 1) the compilation does not work on
>> my computer (neither .tex -> .dvi nor .tex -> .pdf), 2) MiKTeX
>> administrative update (copystart_admin.exe) ends with the error message:
>>
>> =============================
>>
>> Sorry, but something went wrong:
>>
>> Too many open files: path = "C:\Program Files\MiKTeX
>> 2.9\tpm\packages\miktex-metapost-bin-2.9.tpm"
>>
>> path = "C:\Program Files\MiKTeX
>> 2.9\tpm\packages\miktex-metapost-bin-2.9.tpm"
>>
>> =============================
>>
>> I have Windows 7 Professional SP1 as my operating system, with the
>> latest Windows updates installed. It is clear that it will be necessary
>> to deinstall and completely reinstall MiKTeX if I still wish to use
>> MiKTeX in future. However, while there are so many error messages from
>> other colleagues I am not sure if it is worth to spend my time for
>> reinstalling MiKTeX.
>>
>> I am not a MiKTeX developer; I am completely uninterested to study the
>> complicated directory structure of MiKTeX software, the role of each
>> .dll library etc. My only wish is to have some usable tool in order to
>> produce scientific papers in my branch. I ask Christian Schenk and all
>> the other relevant people to restore on all servers the MiKTeX version
>> which was operative until the beginning of May 2016. Please launch only
>> such MiKTeX updates which have been thoroughly tested.
>>
>> Best wishes,
>>
>> Juris Freimanis
>>
>> On 2016.05.31. 12:08, Ulrike Fischer wrote:
>>> Am Tue, 31 May 2016 09:20:02 +0200 schrieb Jan Přikryl:
>>>
>>>> since the last "major" MiKTeX update approximately two weeks ago, MiKTeX
>>>> fails to completely install a new package (be it on-the-fly or through
>>>> MiKTeX package manager). On manual install it always complains about
>>>> UserInstall: C:\Users\Jan Přikryl\AppData\Roaming\MiKTeX\2.9
>>> I saw a similar error message on tex.sx and had a discussion about
>>> it in the chat, but nothing definitive was found.
>>>
>>> He also made a bug report but the error couldn't be reproduced:
>>>
>>> https://sourceforge.net/p/miktex/bugs/2468/
>>>
>>> Like you this user had a non-ascii chars in the user name, so it is
>>> possible that this leads to the failure of fc-cache to write the
>>> cache.
>>>
>>>
>>>
>>
>> ------------------------------------------------------------------------------
>> What NetFlow Analyzer can do for you? Monitors network bandwidth and
>> traffic
>> patterns at an interface-level. Reveals which users, apps, and protocols
>> are
>> consuming the most bandwidth. Provides multi-vendor support for NetFlow,
>> J-Flow, sFlow and other flows. Make informed decisions using capacity
>> planning reports.
>> https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
>> _______________________________________________
>> Q: How can I leave the mailing list?
>> A: See http://docs.miktex.org/faq/support.html#leavingml
>


------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are
consuming the most bandwidth. Provides multi-vendor support for NetFlow,
J-Flow, sFlow and other flows. Make informed decisions using capacity
planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
_______________________________________________
Q: How can I leave the mailing list?
A: See http://docs.miktex.org/faq/support.html#leavingml
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: fc-cache fails after recent MikTeX update

Jan Přikryl
In reply to this post by Ulrike Fischer-2
Dear Ulrike,

Dne Út, 31. květen 2016 v 12:25 h uživatel Ulrike Fischer napsal:
> You need to start debug view with admin rights to get output (the
> tests can then be done in some cmd window with user rights).

Sorry for not being specific enough: I did run the Sysinternals tools as
administrator, and I ran the fc-cache from console with administrative
rights. Being aware of some issues with anti virus software, I have
switched off the NOD32 as well. But, as I said, no output about files in
the console (when I set MIKTEX_TRACE=env, I get some environment
variables listed in the Debug View, so it the trace itself seems to
work). The system is Czech Win 7 SP1 64bit, but the installed MiKTeX is
32bit.

> > Any clue what I can do, short of reinstalling MiKTeX?
>
> If you reinstall miktex you should setup the user roots  so that
> they don't contain non-ascii chars in the name. But it would be
> naturally better to get at the source of the problem. Add a comment
> to the bug report that you are affected too.

That is exactly the reason why I am reluctant to reinstall. In fact, at
this moment this is more a nuisance than a serious error: I have to
install all new packages manually and live with the fact that some font
caches and font maps may be incorrect.

I still have a PC with 2.9.5840 at my other work and it seems to work
okay. My user name at that computer is "prikryl", without any accents,
and I am tempted to test the upgrade tomorrow. In case that the update
is smooth, it would be another indication that the cause of the problem
lays somewhere in handling accented paths on Windows ...

I am of course willing to try a noisy debug version of fc-cache or other
MiKTeX tools, I just do not have enough time to try to build MiKTeX
2.9.5900 from sources at the moment (my build chain is still based on
Visual Studio 2010, I do not have all the prerequisites and I have a few
papers to finish writing ...)

--
  Jan Přikryl
  [hidden email]

------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are
consuming the most bandwidth. Provides multi-vendor support for NetFlow,
J-Flow, sFlow and other flows. Make informed decisions using capacity
planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
_______________________________________________
Q: How can I leave the mailing list?
A: See http://docs.miktex.org/faq/support.html#leavingml
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Too many open files (was fc-cache fails after recent MikTeX update)

Ulrike Fischer-2
In reply to this post by Juris Freimanis
Am Tue, 31 May 2016 14:25:43 +0300 schrieb Juris Freimanis:

> I have not hijacked anything. I received e-mail from Ulrike Fischer
> today at 12:08 East European Daylight Saving Time (= GMT + 3h). Ulrike's
> e-mail was addressed to the whole list
> [hidden email], including myself. Your e-mail was
> automatically cited In that Ulrike's e-mail. I wrote my e-mail as a
> reply to Ulrike's e-mail,

My mail was an answer to Jan's question. When you used it to write
your own question a reference to my mail was added in the header.
This means that in a threaded view (as I'm using) your mail is part
of the thread.

You can see how this looks e.g. here
http://news.gmane.org/gmane.comp.tex.miktex

This is called "hijacking" a thread and you should better not do it.
Apart from the wrong threading it also bears the danger that your
mail is burried deep in some old thread. If you have a new question
create a new mail and don't use an old one to answer.


--
Ulrike Fischer
http://www.troubleshooting-tex.de/


------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are
consuming the most bandwidth. Provides multi-vendor support for NetFlow,
J-Flow, sFlow and other flows. Make informed decisions using capacity
planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
_______________________________________________
Q: How can I leave the mailing list?
A: See http://docs.miktex.org/faq/support.html#leavingml
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: thread hijacking (was: Too many open files)

AlanIsaac
> Am Tue, 31 May 2016 14:25:43 +0300 schrieb Juris Freimanis:
>> I have not hijacked anything. I received e-mail from Ulrike Fischer
>> today at 12:08 East European Daylight Saving Time (= GMT + 3h). Ulrike's
>> e-mail was addressed to the whole list
>> [hidden email], including myself. Your e-mail was
>> automatically cited In that Ulrike's e-mail. I wrote my e-mail as a
>> reply to Ulrike's e-mail,

On 5/31/2016 8:11 AM, Ulrike Fischer wrote:

> My mail was an answer to Jan's question. When you used it to write
> your own question a reference to my mail was added in the header.
> This means that in a threaded view (as I'm using) your mail is part
> of the thread.
>
> You can see how this looks e.g. here
> http://news.gmane.org/gmane.comp.tex.miktex
>
> This is called "hijacking" a thread and you should better not do it.
> Apart from the wrong threading it also bears the danger that your
> mail is burried deep in some old thread. If you have a new question
> create a new mail and don't use an old one to answer.



Illustrated:
http://www.urbandictionary.com/define.php?term=Thread%20Hijacking

Cheers,
Alan

------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are
consuming the most bandwidth. Provides multi-vendor support for NetFlow,
J-Flow, sFlow and other flows. Make informed decisions using capacity
planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
_______________________________________________
Q: How can I leave the mailing list?
A: See http://docs.miktex.org/faq/support.html#leavingml
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: thread hijacking

Juris Freimanis
Sorry, I did not know the term "thread hijacking", nor that at least
some people regard this as bad practice. I had not problems nor
complaints from anybody about this until this moment. I work on PCs
since 1991.

I shall avoid replying on e-mails written by other people in the forum
[hidden email] . I do not see any necessity to
change my behaviour in this context in other conversations, because the
same behaviour is characteristic for almost all my scientific and
non-scientific correspondents.

Juris Freimanis


On 2016.05.31. 15:37, Alan Isaac wrote:

>> Am Tue, 31 May 2016 14:25:43 +0300 schrieb Juris Freimanis:
>>> I have not hijacked anything. I received e-mail from Ulrike Fischer
>>> today at 12:08 East European Daylight Saving Time (= GMT + 3h). Ulrike's
>>> e-mail was addressed to the whole list
>>> [hidden email], including myself. Your e-mail was
>>> automatically cited In that Ulrike's e-mail. I wrote my e-mail as a
>>> reply to Ulrike's e-mail,
> On 5/31/2016 8:11 AM, Ulrike Fischer wrote:
>> My mail was an answer to Jan's question. When you used it to write
>> your own question a reference to my mail was added in the header.
>> This means that in a threaded view (as I'm using) your mail is part
>> of the thread.
>>
>> You can see how this looks e.g. here
>> http://news.gmane.org/gmane.comp.tex.miktex
>>
>> This is called "hijacking" a thread and you should better not do it.
>> Apart from the wrong threading it also bears the danger that your
>> mail is burried deep in some old thread. If you have a new question
>> create a new mail and don't use an old one to answer.
>
>
> Illustrated:
> http://www.urbandictionary.com/define.php?term=Thread%20Hijacking
>
> Cheers,
> Alan
>
> ------------------------------------------------------------------------------
> What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
> patterns at an interface-level. Reveals which users, apps, and protocols are
> consuming the most bandwidth. Provides multi-vendor support for NetFlow,
> J-Flow, sFlow and other flows. Make informed decisions using capacity
> planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
> _______________________________________________
> Q: How can I leave the mailing list?
> A: See http://docs.miktex.org/faq/support.html#leavingml


------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are
consuming the most bandwidth. Provides multi-vendor support for NetFlow,
J-Flow, sFlow and other flows. Make informed decisions using capacity
planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
_______________________________________________
Q: How can I leave the mailing list?
A: See http://docs.miktex.org/faq/support.html#leavingml
Loading...