Package: tracker
Version: 0.6.91-1
Severity: normal

tracker-indexer is eating lots of CPU cycles... and here's strace output for it :

[pid  3903] stat64("/etc/localtime", {st_mode=S_IFREG|0644, st_size=2945, ...}) = 0
[pid  3903] write(12, "\n06 avr 2009, 08:00:40: Tracker-W"..., 92) = 92
[pid  3903] write(2, "\n(tracker-indexer:3903): Tracker-"..., 93) = 93
[pid  3903] clock_gettime(CLOCK_MONOTONIC, {92444, 998761870}) = 0
[pid  3903] gettimeofday({1238997640, 678650}, NULL) = 0
[pid  3903] time(NULL)                  = 1238997640
[pid  3903] stat64("/etc/localtime", {st_mode=S_IFREG|0644, st_size=2945, ...}) = 0
[pid  3903] write(12, "\n06 avr 2009, 08:00:40: Tracker-W"..., 94) = 94
[pid  3903] write(2, "\n(tracker-indexer:3903): Tracker-"..., 95) = 95
[pid  3903] clock_gettime(CLOCK_MONOTONIC, {92444, 999125698}) = 0
[pid  3903] gettimeofday({1238997640, 679014}, NULL) = 0
[pid  3903] time(NULL)                  = 1238997640
[pid  3903] stat64("/etc/localtime", {st_mode=S_IFREG|0644, st_size=2945, ...}) = 0
[pid  3903] write(12, "\n06 avr 2009, 08:00:40: Tracker-W"..., 97) = 97
[pid  3903] write(2, "\n(tracker-indexer:3903): Tracker-"..., 98) = 98
[pid  3903] clock_gettime(CLOCK_MONOTONIC, {92444, 999488288}) = 0
[pid  3903] gettimeofday({1238997640, 679376}, NULL) = 0
[pid  3903] time(NULL)                  = 1238997640
[pid  3903] stat64("/etc/localtime", {st_mode=S_IFREG|0644, st_size=2945, ...}) = 0
[pid  3903] write(12, "\n06 avr 2009, 08:00:40: Tracker-W"..., 95) = 95
[pid  3903] write(2, "\n(tracker-indexer:3903): Tracker-"..., 96) = 96
[pid  3903] clock_gettime(CLOCK_MONOTONIC, {92444, 999851406}) = 0
[pid  3903] gettimeofday({1238997640, 679739}, NULL) = 0
[pid  3903] time(NULL)                  = 1238997640

almost continuously doing the same things... dunno why this repeats ad nauseam... but I suspect it's busy checking time and logging whatever it does.... in no real purpose... of course these are only system calls and it might do other seems which justify the 80% use of the CPU in top.

Ain't there any programming problem ?

Just my 2 cents,

Best regards,

P.S.: dear maintainer... could you acknowledge problems reported to the BTS about what you packaged ? ... do you intend to respond to users some day ? Many thanks in advance.

Olivier Berger wrote:

I do answer to proper bug reports, yes.

Michael Biebl <biebl@debian.org> writes:

What was missing in my reportbug-submitted bug ?

Do you need more details ?

Oh, and btw, I was mainly referring to a previous problem in #521380 which
was pretty annoying too... and which seem to have been unnoticed so
far... at least from what I can tell.
Same stands for #467270 which got no more love even if there have been
a few more people complaining :-(

It's not so obvious if this is Debian specific, and if upstream is
aware (at least) either.

I hope we'll be able to help identify the problems here anyway.

Thanks for your response : I now know you're caring for tracker at least.

Regards,

Hi,

could you please upgrade to the latest 0.8 version from unstable/testing and try
to reproduce the problem there.

Development of the 0.6.9x series is discontinued and our focus should be on the
0.8 branch, which had some major architectural changes.

Thanks,
Michael