This is the mail archive of the systemtap@sourceware.org mailing list for the systemtap project.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
Other format: | [Raw text] |
Hi,I used bulk-mode before to collect user-space call-graph of web servers using the script https://github.com/openresty/nginx-systemtap-toolkit/blob/master/sample-bt
stap generates bunch of files, only one has data, which is consistent with our settings --
we set CPU affinity such that each web-server is bound to a particular CPU.The problem and the only problem is that the merged resulting file is suspicious as the sample-count seems to be significantly less than the one we collected using
non-bulk mode. Do we need to raise the sampling frequency for bulk mode? Thanks Shuxin On 01/07/2016 10:43 AM, Frank Ch. Eigler wrote:
Hi -[...] ERROR: context (size 66144) allocation failed WARNING: /usr/local/systemtap/bin/staprun exited with status: 1 [...]The context structure is allocated per-cpu, so for high-cpu-count boxes can be a problem. You must be running stap v2.4 or older, based on that message. With newer stap, the context sizes are sometimes dramatically smaller for the same script. - FChE
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |