Hi again Do you have an NFS mounted into some subfolder of ~ or is your ~ on an NFS?
I can reproduce the same issue on my desktop -- it works if I only include "simple" subfolders of ~, but it fails when trying to index the whole ~. mfg Tobias On 11 August 2017 at 11:10, Mark Dixon <[email protected]> wrote: > I've stopped in now, would have to set it going again. CPU is not the > issue though, it's usage is minimal, but the system is slow to the point of > unresponsiveness. It's kinda weird, it looks idle in top but it feels > overloaded. > > That's why I started looking at file handles etc. > > Mark > > > On 11 Aug 2017, at 09:08, Tobias C. Berner <[email protected]> wrote: > > Hi Mark > > what's the output of 'balooctl stauts'? Is it still doing the initial > indexing -- that's the one that causes a lot of cpu usage. Afterwards it > should calm down.... theoretically :D > > mfg Tobias > > On 10 August 2017 at 22:12, Mark Dixon <[email protected]> wrote: > >> Hi all, >> >> I'm having problems with Baloo again. I'm not sure exactly what the >> requirements are for it, but I must be doing something wrong. It >> essentially renders the system unusable. >> >> I finally caught it in this state - CPU is minimal, and memory isn't too >> bad (for this box) - baloo_file was using 1.5G. What surprised me was the >> output of lsof: >> >> baloo_fil 1808 mark *881r VREG 110,3406168081 >> 20453 44020 /usr/home/mark/cudaminer/CudaMiner/compat/jansson/value.c >> baloo_fil 1808 mark *882r VREG >> 110,3406168081 306 44006 /usr/home/mark/cudaminer/CudaM >> iner/compat/jansson/Makefile.am >> baloo_fil 1808 mark *883r VREG 110,3406168081 >> 14123 30810 /usr/home/mark/cudaminer/CudaM >> iner/compat/jansson/Makefile >> baloo_fil 1808 mark *884r VREG 110,3406168081 >> 2225 44015 /usr/home/mark/cudaminer/CudaM >> iner/compat/jansson/strbuffer.c >> baloo_fil 1808 mark *885r VREG 110,3406168081 >> 243238 44360 /usr/home/mark/cudaminer/CudaM >> iner/compat/jansson/libjansson.a >> baloo_fil 1808 mark *886r VREG 110,3406168081 >> 11784 44356 /usr/home/mark/cudaminer/CudaMiner/compat/jansson/utf.o >> baloo_fil 1808 mark *887r VREG 110,3406168081 >> 66568 44352 /usr/home/mark/cudaminer/CudaMiner/compat/jansson/load.o >> baloo_fil 1808 mark *888r VREG 110,3406168081 >> 1439 44013 /usr/home/mark/cudaminer/CudaMiner/compat/jansson/jansson_ >> private.h >> baloo_fil 1808 mark *889r VREG 110,3406168081 >> 80232 44358 /usr/home/mark/cudaminer/CudaMiner/compat/jansson/value.o >> baloo_fil 1808 mark *890r VREG 110,3406168081 >> 33800 44350 /usr/home/mark/cudaminer/CudaM >> iner/compat/jansson/hashtable.o >> baloo_fil 1808 mark *891r VREG 110,3406168081 >> 13152 44009 /usr/home/mark/cudaminer/CudaMiner/compat/jansson/dump.c >> baloo_fil 1808 mark *892r VREG >> 110,3406168081 897 44016 /usr/home/mark/cudaminer/CudaM >> iner/compat/jansson/strbuffer.h >> baloo_fil 1808 mark *893r VREG 110,3406168081 >> 10368 44354 /usr/home/mark/cudaminer/CudaM >> iner/compat/jansson/strbuffer.o >> baloo_fil 1808 mark *894r VREG 110,3406168081 >> 6214 44011 /usr/home/mark/cudaminer/CudaM >> iner/compat/jansson/hashtable.h >> >> etc... >> >> The number of lines was quite impressive though: >> >> $ lsof | grep 1808 | wc -l >> 983978 >> >> What is it doing with all these files? Is this likely the cause of the >> unresponsive system? >> >> Mark >> >> >> >> >> >
