fche changed the topic of #systemtap to: http://sourceware.org/systemtap; email systemtap@sourceware.org if answers here not timely, conversations may be logged
hchiramm has quit [Ping timeout: 244 seconds]
pmuldoon has quit [Ping timeout: 264 seconds]
hpt has joined #systemtap
pmuldoon has joined #systemtap
scox has joined #systemtap
hchiramm_ has quit [Ping timeout: 250 seconds]
hchiramm_ has joined #systemtap
p53ud0nym has quit [Remote host closed the connection]
pmuldoon has quit [Remote host closed the connection]
msekleta has joined #systemtap
<msekleta>
hello, does anyone know how to figure out to which cgroups does given PID belongs to in context of systemtap probe? Probe point I am using right now is signal.send and from there it seems impossible to get information I need.
<fche>
msekleta, not sure we have that mapping easily available
<fche>
but you can get intra-container pid vs. host-level pid
<fche>
see ns_pid() vs pid()
<msekleta>
fche: PID sending signal is not running in pid namespace
<fche>
might be able to trap receipt of signal and look there?
<msekleta>
I wanted to obtain cgroup path because I wanted figure which system service PID sender belongs to
<fche>
any idea whether such a path is available within the kernel from the task_struct ?
<msekleta>
dunno
<msekleta>
checking now
<fche>
another way could be to have stap track pid/namespace transitions with auxiliary probes, and store current mappings in systemtap-side lookup tables
ravi_ has quit [Ping timeout: 250 seconds]
<msekleta>
fche: in task_struct there is struct css_set __rcu *cgroups
<msekleta>
that seems like a good starting point
<msekleta>
css_set then contains struct cgroup_subsys_state *subsys[CGROUP_SUBSYS_COUNT];
<msekleta>
from comments in the code it seems like cgroup_subsys_state is an object containing state information for given cgroup in a given subsystem (controller)
mbenitez has joined #systemtap
mbenitez has quit [Changing host]
mbenitez has joined #systemtap
<msekleta>
and cgroup_subsys_state has finally pointer to struct cgroup
pmuldoon has joined #systemtap
<msekleta>
also there is task_cgroup function which does basically then same
scox has joined #systemtap
srikar is now known as srikar_away
<msekleta>
fche: pr_cont_cgroup_path is a function that directly (via printk) prints cgroup path
pmuldoon has quit [Remote host closed the connection]
wcohen has quit [Ping timeout: 272 seconds]
ph7 has quit [Read error: Connection timed out]
ph7 has joined #systemtap
ph7 has quit [Quit: Leaving.]
tromey has joined #systemtap
ph7 has joined #systemtap
brolley has joined #systemtap
pmuldoon has joined #systemtap
<fche>
msekleta, ok, good poiners
drsmith_away is now known as drsmith
ph7 has quit [Quit: Leaving.]
ph7 has joined #systemtap
mbenitez has quit [Quit: To office]
ego_ has joined #systemtap
ph7 has quit [Quit: Leaving.]
wcohen has joined #systemtap
ph7 has joined #systemtap
ph7 has quit [Quit: Leaving.]
srikar_away is now known as srikar
hkshaw has quit [Ping timeout: 252 seconds]
mbenitez has joined #systemtap
hkshaw has joined #systemtap
pmuldoon has quit [Remote host closed the connection]
hkshaw has quit [Ping timeout: 260 seconds]
ph7 has joined #systemtap
ph7 has quit [Quit: Leaving.]
pmuldoon has joined #systemtap
pmuldoon has quit [Read error: Connection reset by peer]
pmuldoon has joined #systemtap
ph7 has joined #systemtap
hkshaw has joined #systemtap
nkambo has quit [Read error: No route to host]
nkambo has joined #systemtap
zodbot has quit [Disconnected by services]
hkshaw has quit [Read error: Connection reset by peer]
hkshaw1 has joined #systemtap
zodbot has joined #systemtap
mjw has quit [Quit: Leaving]
ph7 has quit [Read error: Connection timed out]
ph7 has joined #systemtap
nkambo1 has joined #systemtap
nkambo has quit [Ping timeout: 252 seconds]
irker177 has joined #systemtap
<irker177>
systemtap: mcermak systemtap.git:refs/heads/master * release-3.0-167-gdadf22f / testsuite/systemtap.stress/tapset_functions.exp: Don't use as_root() to run stap within tapset_functions.exp http://tinyurl.com/j5pbzen
ego_ has quit [Ping timeout: 240 seconds]
detter has joined #systemtap
hchiramm has quit [Ping timeout: 264 seconds]
<irker177>
systemtap: jistone systemtap.git:refs/heads/master * release-3.0-168-g91bfb36 / testsuite/systemtap.examples/memory/pfaults.stp: pfaults.stp: correct the fault_entry_time index http://tinyurl.com/grs2pzv
ph7 has quit [Quit: Leaving.]
srikar is now known as srikar_away
<detter>
jstone, I am now running on Ubuntu 14.04 and now I get an intermittent crash: ERROR: user string copy fault -14 at 00007fb50551d8c3 near identifier 'user_string_n' at /usr/share/systemtap/tapset/uconversions.stp:120:10
<jistone>
detter, when this happens with a valid-looking pointer, usually it means the program hasn't actually paged-in that memory yet
<jistone>
fche had some library hack to force programs to mlockall, which is crude but effective
<jistone>
detter, it might also help to limit your probe to specific processes, in case it happens to be something else on the system that's faulting you
<jistone>
if (execname() ~= "foo) ...
<detter>
ok, it isn't a huge issue, I'm just getting it once in a while so it's more of just a nuisance
<jistone>
or rather, =~
<jistone>
I always forget
<detter>
jistone, thanks! =)
<fche>
thanks josh
mjw has joined #systemtap
hkshaw1 has quit [Ping timeout: 276 seconds]
drsmith has left #systemtap [#systemtap]
msekleta has quit [Quit: WeeChat 1.5]
_whitelogger has joined #systemtap
brolley has left #systemtap [#systemtap]
wcohen has quit [Ping timeout: 272 seconds]
mjw has quit [Quit: Leaving]
gila has joined #systemtap
scox has quit [Ping timeout: 244 seconds]
detter has quit [Quit: Leaving]
detter has joined #systemtap
detter has quit [Remote host closed the connection]
detter has joined #systemtap
detter has quit [Client Quit]
detter has joined #systemtap
drsmith has left #systemtap [#systemtap]
mbenitez has quit [Quit: Leaving]
gila has quit [Quit: My Mac Pro has gone to sleep. ZZZzzz…]