Re: Review Request 33040: Expose qdisc statistics from libnl

2015-04-24 Thread Cong Wang
/r/33040/#comment131945 This fixes a real bug, should be a separated patch. - Cong Wang On April 24, 2015, 7:28 p.m., Paul Brett wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r

Review Request 33759: Add schedule delays to cpu statistics

2015-05-01 Thread Cong Wang
d0932b04e3825abb6173efe0d1aee199aa356932 Diff: https://reviews.apache.org/r/33759/diff/ Testing --- Manual tests on local host Thanks, Cong Wang

Re: Review Request 31504: Add a basic filter to match all packets

2015-05-05 Thread Cong Wang
/routing/filter/basic.hpp PRE-CREATION src/linux/routing/filter/basic.cpp PRE-CREATION src/tests/routing_tests.cpp ce583b59bf9fb2ef855aa82ab6083ea11b138e55 Diff: https://reviews.apache.org/r/31504/diff/ Testing --- Run the testcase. Thanks, Cong Wang

Re: Review Request 31505: Add flow classifiers for fq_codel on egress

2015-05-11 Thread Cong Wang
/ Testing --- Manually start two mesos containers with netperf running side. Thanks, Cong Wang

Re: Review Request 31505: Add flow classifiers for fq_codel on egress

2015-05-14 Thread Cong Wang
--- On May 12, 2015, 12:14 a.m., Cong Wang wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/31505/ --- (Updated May 12

Re: Review Request 31505: Add flow classifiers for fq_codel on egress

2015-05-14 Thread Cong Wang
--- On May 12, 2015, 12:14 a.m., Cong Wang wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/31505

Re: Review Request 34426: Extend the Traffic Control queueing models to allow dynamically allocated Handles, extend API to allow searching by handle (or parent).

2015-05-19 Thread Cong Wang
, it reads like a handle of veth. Please rename it to 'egressHandle'. - Cong Wang On May 19, 2015, 7:55 p.m., Paul Brett wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/34426

Re: Review Request 31505: Add flow classifiers for fq_codel on egress

2015-05-16 Thread Cong Wang
with netperf running side. Thanks, Cong Wang

Review Request 35080: Fix comments in routing_tests.cpp

2015-06-04 Thread Cong Wang
Description --- s/fq_codel filter/fq_codel qdisc/ s/ingress filter/ingress qdisc/ s/non-existant/nonexistent/ Diffs - src/tests/routing_tests.cpp 96be25fc31a0024642854f65fc3410e05b4576b7 Diff: https://reviews.apache.org/r/35080/diff/ Testing --- make check Thanks, Cong Wang

Re: Review Request 35225: Add HTB queueing discipline wrapper class

2015-06-08 Thread Cong Wang
/35225/#comment139365 These params are for fq_codel, doesn't apply for htb. - Cong Wang On June 8, 2015, 9:46 p.m., Paul Brett wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/35225

Re: Review Request 35152: Document and consolidate qdisc handles

2015-06-05 Thread Cong Wang
/35152/#comment138935 Ditto, the fq_codel here is for traffic isolation and reducing buffer-bloat. - Cong Wang On June 5, 2015, 8:38 p.m., Paul Brett wrote: --- This is an automatically generated e-mail. To reply, visit: https

Re: Review Request 35152: Document and consolidate qdisc handles

2015-06-05 Thread Cong Wang
/#comment139004 s/CLASS_ID/HTB_CLASS_ID/ ? - Cong Wang On June 5, 2015, 11:55 p.m., Paul Brett wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/35152

Re: Review Request 35229: Report per-container metrics for network bandwidth throttling to the slave.

2015-06-09 Thread Cong Wang
, 10:26 p.m.) Review request for mesos, Chi Zhang, Ian Downes, Jie Yu, and Cong Wang. Bugs: MESOS-2332 https://issues.apache.org/jira/browse/MESOS-2332 Repository: mesos Description --- Report per-container metrics for network bandwidth throttling to the slave

Re: Review Request 35225: Add HTB queueing discipline wrapper class

2015-06-09 Thread Cong Wang
/35225/#comment139549 s/if a queueing discipline already exists/if an htb queueing discipline already exists/ ?? src/linux/routing/queueing/htb.hpp https://reviews.apache.org/r/35225/#comment139548 s/Return/Returns/ - Cong Wang On June 8, 2015, 10:58 p.m., Paul Brett wrote

Review Request 35332: Add a test case for egress flow ID recovery

2015-06-10 Thread Cong Wang
--- make check Thanks, Cong Wang

Review Request 35331: Add a test case for egress flow ID

2015-06-10 Thread Cong Wang
/network/port_mapping.cpp 432b05ce5a99c8239fafc47a6b65d46a0fbac26e src/tests/port_mapping_tests.cpp f8372df74cd71df37de4a2438069ef0ea8878512 Diff: https://reviews.apache.org/r/35331/diff/ Testing --- make check Thanks, Cong Wang

Re: Review Request 31505: Add flow classifiers for fq_codel on egress

2015-06-01 Thread Cong Wang
/port_mapping.cpp 843e52d6f9923d0ee0a0297cd5c464b8b72f5de3 Diff: https://reviews.apache.org/r/31505/diff/ Testing --- Manually start two mesos containers with netperf running side. Thanks, Cong Wang

Re: Review Request 31505: Add flow classifiers for fq_codel on egress

2015-05-27 Thread Cong Wang
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/31505/#review84998 --- On May 26, 2015, 8:41 p.m., Cong Wang wrote

Re: Review Request 31505: Add flow classifiers for fq_codel on egress

2015-05-27 Thread Cong Wang
: Ping. Did you see these two comments? Seems that the new diff hasn't resolved these two issues. Cong Wang wrote: Yes. 1) I think we all agree on not adding a new flag in case of inconsistent behavior; 2) We already have a test case, FqCodelClassifier. Jie Yu wrote: For 1), I think

Re: Review Request 31505: Add flow classifiers for fq_codel on egress

2015-05-27 Thread Cong Wang
: Ping. Did you see these two comments? Seems that the new diff hasn't resolved these two issues. Cong Wang wrote: Yes. 1) I think we all agree on not adding a new flag in case of inconsistent behavior; 2) We already have a test case, FqCodelClassifier. Jie Yu wrote: For 1), I think

Re: Review Request 31505: Add flow classifiers for fq_codel on egress

2015-06-02 Thread Cong Wang
on. Cong Wang wrote: No, https://reviews.apache.org/r/31503/ this one should be the first commit requires 3.2.26. Chi Zhang wrote: hmm looks like both are committed? could you submit a fix for that please? Done: https://issues.apache.org/jira/browse/MESOS-2803 . Feel free to take

Re: Review Request 31505: Add flow classifiers for fq_codel on egress

2015-06-01 Thread Cong Wang
--- On June 1, 2015, 10:45 p.m., Cong Wang wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/31505

Re: Review Request 34426: Report per-container metrics for network bandwidth throttling

2015-06-01 Thread Cong Wang
On June 1, 2015, 6:44 p.m., Cong Wang wrote: src/linux/routing/queueing/internal.hpp, line 284 https://reviews.apache.org/r/34426/diff/8/?file=975301#file975301line284 The lastest mesos code already relies on libnl 3.2.26, so not sure how much sense it makes to keep the backward

Re: Review Request 34321: Merge class Handle which is duplicated between filter/handle and queueing/handle.

2015-05-22 Thread Cong Wang
On May 22, 2015, 5:28 a.m., Cong Wang wrote: src/linux/routing/handle.hpp, line 118 https://reviews.apache.org/r/34321/diff/7/?file=968966#file968966line118 Same comments near the definitions in handle.cpp, duplicated? Paul Brett wrote: Once for the declaration, second

Re: Review Request 31505: Add flow classifiers for fq_codel on egress

2015-05-26 Thread Cong Wang
49e983edab598e2ac487bb488fdd12840a9e7dfc Diff: https://reviews.apache.org/r/31505/diff/ Testing --- Manually start two mesos containers with netperf running side. Thanks, Cong Wang

Re: Review Request 31505: Add flow classifiers for fq_codel on egress

2015-05-26 Thread Cong Wang
--- On May 26, 2015, 8:41 p.m., Cong Wang wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/31505/ --- (Updated May 26, 2015, 8:41 p.m

Re: Review Request 37399: Ignore no statistics condition for containers with no qdisc.

2015-08-12 Thread Cong Wang
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/37399/#review95220 --- Ship it! Ship It! - Cong Wang On Aug. 12, 2015, 9:09 p.m., Paul

Re: Review Request 37541: Add TraceEvent API

2015-08-17 Thread Cong Wang
--- make check Thanks, Cong Wang

Re: Review Request 37416: Perf supported() should be based on the version of perf, not the version of the kernel.

2015-08-24 Thread Cong Wang
/#comment151460 This is not expected, right? src/linux/perf.cpp (line 477) https://reviews.apache.org/r/37416/#comment151461 Make it complete, Perf version is not available. - Cong Wang On Aug. 21, 2015, 6:46 p.m., Paul Brett wrote

Re: Review Request 37416: Perf supported() should be based on the version of perf, not the version of the kernel.

2015-08-24 Thread Cong Wang
On Aug. 24, 2015, 5:53 p.m., Cong Wang wrote: src/linux/perf.cpp, line 418 https://reviews.apache.org/r/37416/diff/5/?file=1045148#file1045148line418 This is not expected, right? Paul Brett wrote: It would be a rare event but not completly unexpected. After all, perf can

Re: Review Request 37466: Update perf tests to including testing the supported perf output formats.

2015-08-24 Thread Cong Wang
://reviews.apache.org/r/37466/#comment151466 I think this should still be in perf.hpp even though it is just for testing. - Cong Wang On Aug. 21, 2015, 4:06 a.m., Paul Brett wrote: --- This is an automatically generated e-mail. To reply, visit

Review Request 37540: Add PerfEvent API

2015-08-17 Thread Cong Wang
cb1a13d0b1754a50f0121bfda522056ff8c3e3c8 src/tests/containerizer/perf_tests.cpp 6b3d70f3e7ea8f59f94e6961491d4e9a730e3334 Diff: https://reviews.apache.org/r/37540/diff/ Testing --- make check Thanks, Cong Wang

Review Request 37541: Add TraceEvent API

2015-08-17 Thread Cong Wang
/perf.cpp cb1a13d0b1754a50f0121bfda522056ff8c3e3c8 src/tests/containerizer/perf_tests.cpp 6b3d70f3e7ea8f59f94e6961491d4e9a730e3334 Diff: https://reviews.apache.org/r/37541/diff/ Testing --- make check Thanks, Cong Wang

Review Request 37101: Remove unused sched API's

2015-08-04 Thread Cong Wang
://reviews.apache.org/r/37101/diff/ Testing --- make Thanks, Cong Wang

Re: Review Request 37147: Timeout perf stat command if it does not complete.

2015-08-05 Thread Cong Wang
$DURATION`. Please state it in description - Cong Wang On Aug. 5, 2015, 11:25 p.m., Paul Brett wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/37147

Re: Review Request 37399: Ignore no statistics condition for containers with no qdisc.

2015-08-12 Thread Cong Wang
889) https://reviews.apache.org/r/37399/#comment149942 Could be just Failed to get htb qdisc statistics on - Cong Wang On Aug. 12, 2015, 4:57 p.m., Paul Brett wrote: --- This is an automatically generated e-mail. To reply, visit

Re: Review Request 37540: Add PerfEvent API

2015-08-24 Thread Cong Wang
b77b61d7048b12cea4586bcf802cbc2ff634331b src/linux/perf.cpp cdc5f8314a875ababf2b17a32873775d808e1c78 src/tests/containerizer/perf_tests.cpp 6b3d70f3e7ea8f59f94e6961491d4e9a730e3334 Diff: https://reviews.apache.org/r/37540/diff/ Testing --- make check Thanks, Cong Wang

Re: Review Request 37101: Remove unused sched API's

2015-08-24 Thread Cong Wang
/sched_tests.cpp 00723d01cd2cc37410d6f9fdd2de080063b7ccd8 Diff: https://reviews.apache.org/r/37101/diff/ Testing --- make Thanks, Cong Wang

Re: Review Request 37541: Add TraceEvent API

2015-08-24 Thread Cong Wang
/linux/perf.hpp b77b61d7048b12cea4586bcf802cbc2ff634331b src/linux/perf.cpp cdc5f8314a875ababf2b17a32873775d808e1c78 src/tests/containerizer/perf_tests.cpp 6b3d70f3e7ea8f59f94e6961491d4e9a730e3334 Diff: https://reviews.apache.org/r/37541/diff/ Testing --- make check Thanks, Cong

Re: Review Request 37541: Add TraceEvent API

2015-08-24 Thread Cong Wang
b77b61d7048b12cea4586bcf802cbc2ff634331b src/linux/perf.cpp cdc5f8314a875ababf2b17a32873775d808e1c78 src/tests/containerizer/perf_tests.cpp 6b3d70f3e7ea8f59f94e6961491d4e9a730e3334 Diff: https://reviews.apache.org/r/37541/diff/ Testing --- make check Thanks, Cong Wang

Re: Review Request 37540: Add PerfEvent API

2015-08-24 Thread Cong Wang
b77b61d7048b12cea4586bcf802cbc2ff634331b src/linux/perf.cpp cdc5f8314a875ababf2b17a32873775d808e1c78 src/tests/containerizer/perf_tests.cpp 6b3d70f3e7ea8f59f94e6961491d4e9a730e3334 Diff: https://reviews.apache.org/r/37540/diff/ Testing --- make check Thanks, Cong Wang

Re: Review Request 39494: Added libnl, libevent and ssl flags to config docs.

2015-10-21 Thread Cong Wang
g/r/39494/#comment161489> Like what you do for libevent, it would be nicer if you can add a URL to libnl3 too, but this is not a big deal... - Cong Wang On Oct. 21, 2015, 4:17 p.m., Greg Mann wrote: > > --- > This is a

Re: Review Request 39494: Added libnl, libevent and ssl flags to config docs.

2015-10-21 Thread Cong Wang
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/39494/#review103436 --- Ship it! Ship It! - Cong Wang On Oct. 21, 2015, 5:28 p.m

Re: Review Request 39416: Document --egress_unique_flow_per_container in docs/configuration.md

2015-10-27 Thread Cong Wang
eviews.apache.org/r/39416/#review104117 --- On Oct. 18, 2015, 12:24 a.m., Cong Wang wrote: > > --- > This is an automatically generated e-mail. To re

Re: Review Request 39490: Always create non-IP egress filters

2015-10-27 Thread Cong Wang
---- On Oct. 20, 2015, 6:57 p.m., Cong Wang wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/39490/ > -

Re: Review Request 39415: Error out when root qdisc already exists

2015-10-27 Thread Cong Wang
utomatically generated e-mail. To reply, visit: https://reviews.apache.org/r/39415/#review104187 --- On Oct. 18, 2015, 12:22 a.m., Cong Wang wrote: > > --- > This is an automati

Re: Review Request 39417: Add --egress_flow_classifier_parent flag

2015-10-28 Thread Cong Wang
an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/39417/#review104119 --- On Oct. 18, 2015, 12:29 a.m., Cong Wang wrote: > >

Re: Review Request 39417: Add --egress_flow_classifier_parent flag

2015-10-28 Thread Cong Wang
- Cong --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/39417/#review104120 --- On Oct. 18, 2015, 12:29 a.m., Cong W

Re: Review Request 39415: Error out when root qdisc already exists

2015-10-28 Thread Cong Wang
, Cong Wang

Re: Review Request 39416: Document --egress_unique_flow_per_container in docs/configuration.md

2015-10-28 Thread Cong Wang
Diff: https://reviews.apache.org/r/39416/diff/ Testing --- None. Thanks, Cong Wang

Re: Review Request 39417: Add --egress_flow_classifier_parent flag

2015-10-28 Thread Cong Wang
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/39417/#review104119 ------- On Oct. 18, 2015, 12:29 a.m., Cong Wang wrote: > > --- > This

Re: Review Request 39417: Add --egress_flow_classifier_parent flag

2015-10-28 Thread Cong Wang
tests, with and without a pre-installed HTB qdisc and classes. Thanks, Cong Wang

Re: Review Request 40211: Add mesos provisioner doc.

2015-11-15 Thread Cong Wang
g/r/40211/#comment165290> to be created and managed... docs/mesos-provisioner.md (line 21) <https://reviews.apache.org/r/40211/#comment165289> s/and download/download/ s/unpact/unpack/ - Cong Wang On Nov. 15, 2015, 7:26 p.m., Timoth

Re: Review Request 38117: Export per container SNMP statistics

2015-10-19 Thread Cong Wang
/flags.hpp 3e93b52a5874f52361d5a9c685499a7032014a73 src/slave/flags.cpp 1bf394ea62fde29caa6705cd5d156eae452adbf2 Diff: https://reviews.apache.org/r/38117/diff/ Testing (updated) --- ./src/mesos-network-helper statistics --eth0_name=X --enable_snmp_statistics --pid= Thanks, Cong

Re: Review Request 39360: Relocate MesosContainerizer specific files to the correct location

2015-10-19 Thread Cong Wang
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/39360/#review103157 --- You probably want to move src/tests/containerizer too? - Cong

Re: Review Request 38117: Export per container SNMP statistics

2015-10-19 Thread Cong Wang
side the loop, its value can't be kept for the next loop, right? We need to at least keep keys across one loop. - Cong --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/38117/#review10

Review Request 39490: Always create non-IP egress filters

2015-10-20 Thread Cong Wang
/39490/diff/ Testing --- Manual tests Thanks, Cong Wang

Re: Review Request 39002: Fixed sed command for Python protos generation.

2015-10-05 Thread Cong Wang
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/39002/#review101499 --- Ship it! Ship It! - Cong Wang On Oct. 5, 2015, 5:48 p.m

Re: Review Request 36281: Document per-container unique egress flows and network queueing statistics.

2015-07-07 Thread Cong Wang
/#comment143969 That is called flow classification and isolation. Please also mention that flow is classified based on port ranges. - Cong Wang On July 7, 2015, 9:54 p.m., Paul Brett wrote: --- This is an automatically generated e-mail

Re: Review Request 39417: Add --egress_flow_classifier_parent flag

2015-11-18 Thread Cong Wang
not present so you can use numify? > > > > if (!strings::startsWith(tokens[0], "0x")) { > > ... > > } > > Cong Wang wrote: > This should work too, but I don't feel it is better than my code. > > Ian Downes wrote: > I reall

Re: Review Request 37894: Renamed cgroups::kill to cgroups::signal.

2015-08-28 Thread Cong Wang
a signal handler... - Cong Wang On Aug. 28, 2015, 4:39 p.m., Joerg Schad wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/37894

Re: Review Request 37894: Renamed cgroups::kill to cgroups::signal.

2015-08-28 Thread Cong Wang
On Aug. 28, 2015, 4:45 p.m., Cong Wang wrote: Why? Everyone knows kill(2) sends a signal while signal(2) installs a signal handler... Jie Yu wrote: IMO, that naming is confusing, and should be 'signal' and 'install'. Joerg Schad wrote: This is actually answering

Re: Review Request 37903: stout: Fix bug in IPNetwork::create() with zero prefix.

2015-08-31 Thread Cong Wang
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/37903/#review97170 --- Ship it! Or we can simply make 'prefix' unsigned. - Cong Wang

Re: Review Request 37540: Add perf event API

2015-09-04 Thread Cong Wang
This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/37540/#review97548 --- On Sept. 2, 2015, 10:16 p.m., Cong Wang wrote: > > -

Re: Review Request 37540: Add perf event API

2015-09-04 Thread Cong Wang
s.apache.org/r/37540/#review97548 --- On Sept. 2, 2015, 10:16 p.m., Cong Wang wrote: > > --- > This is an automatically generated e-mail. To reply, v

Re: Review Request 37540: Add perf event API

2015-09-04 Thread Cong Wang
his future? finalize(). - Cong --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/37540/#review97548 --- On Sept. 2,

Re: Review Request 38117: Export per container SNMP statistics

2015-09-08 Thread Cong Wang
b8335aa585376d27b71897f8cbaefaa12f4b3a5c src/slave/flags.cpp 7539441c685828027db07173e62a4e5fc1e8b54d Diff: https://reviews.apache.org/r/38117/diff/ Testing --- Manual tests Thanks, Cong Wang

Re: Review Request 37541: Add TraceEvent API

2015-09-02 Thread Cong Wang
Diff: https://reviews.apache.org/r/37541/diff/ Testing --- make check Thanks, Cong Wang

Review Request 38074: Calculate schedule latency with trace events

2015-09-02 Thread Cong Wang
b36710d6d7a7250bc071a57310a2d54bfb3bc624 Diff: https://reviews.apache.org/r/38074/diff/ Testing --- manual tests Thanks, Cong Wang

Review Request 38073: Avoid parsing perf events twice

2015-09-02 Thread Cong Wang
src/slave/containerizer/isolators/cgroups/perf_event.cpp 8c3018de8e77a2a00c90559a995eae2b3678e42f Diff: https://reviews.apache.org/r/38073/diff/ Testing --- make check Thanks, Cong Wang

Re: Review Request 37540: Add perf event API

2015-09-02 Thread Cong Wang
, Cong Wang

Re: Review Request 37541: Add trace event API

2015-09-02 Thread Cong Wang
8c29ca3dc2a8f6d1fb0f430482190f87f08eb4e8 Diff: https://reviews.apache.org/r/37541/diff/ Testing --- make check Thanks, Cong Wang

Re: Review Request 37540: Add PerfEvent API

2015-09-02 Thread Cong Wang
c5630118f4858b1a805f25a2db7a24ca0989 src/linux/perf.cpp bb8b5918e4c8f4379784d4fa18c3b4709696e1ea src/tests/containerizer/perf_tests.cpp 8c29ca3dc2a8f6d1fb0f430482190f87f08eb4e8 Diff: https://reviews.apache.org/r/37540/diff/ Testing --- make check Thanks, Cong Wang

Re: Review Request 38074: Calculate schedule latency with trace events

2015-09-02 Thread Cong Wang
e56738e2dfd6593ef8f093687919da287af78f77 src/slave/flags.cpp b36710d6d7a7250bc071a57310a2d54bfb3bc624 Diff: https://reviews.apache.org/r/38074/diff/ Testing --- manual tests Thanks, Cong Wang

Re: Review Request 37541: Add trace event API

2015-09-04 Thread Cong Wang
bef475e4b573f02b69f2c4dd6d1c6a6e10f34c38 Diff: https://reviews.apache.org/r/37541/diff/ Testing --- make check Thanks, Cong Wang

Re: Review Request 38074: Calculate schedule latency with trace events

2015-09-04 Thread Cong Wang
b8335aa585376d27b71897f8cbaefaa12f4b3a5c src/slave/flags.cpp 7539441c685828027db07173e62a4e5fc1e8b54d Diff: https://reviews.apache.org/r/38074/diff/ Testing --- manual tests Thanks, Cong Wang

Re: Review Request 37540: Add perf event API

2015-09-04 Thread Cong Wang
/linux/perf.hpp d10968ca670eb516bae08385fd0ddde8e8ad83b5 src/linux/perf.cpp 0011482cf9d920485728798518d32af0e9627724 src/tests/containerizer/perf_tests.cpp bef475e4b573f02b69f2c4dd6d1c6a6e10f34c38 Diff: https://reviews.apache.org/r/37540/diff/ Testing --- make check Thanks, Cong

Re: Review Request 38233: os: add swap information to memory().

2015-09-09 Thread Cong Wang
(line 757) <https://reviews.apache.org/r/38233/#comment154664> Bytes(usage.xsu_total * pageSize)? - Cong Wang On Sept. 9, 2015, 9:32 p.m., Chi Zhang wrote: > > --- > This is an automatically generated e-mail. To reply,

Re: Review Request 38074: Calculate schedule latency with trace events

2015-09-09 Thread Cong Wang
----- On Sept. 4, 2015, 11:16 p.m., Cong Wang wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/38074/ > ---

Re: Review Request 38382: Fixed the perf event isolator to continue sampling in the presence of failures.

2015-09-14 Thread Cong Wang
) <https://reviews.apache.org/r/38382/#comment155717> s/sampling will be halted// ? - Cong Wang On Sept. 14, 2015, 10:57 p.m., Ben Mahler wrote: > > --- > This is an automatically generated e-mail. To reply,

Re: Review Request 38379: Handle empty set of cgroups as a no-op in perf::sample.

2015-09-14 Thread Cong Wang
calls it with non-empty cgroups. Also, you probably want to check for events.empty() too while you are on it. - Cong Wang On Sept. 14, 2015, 10:57 p.m., Ben Mahler wrote: > > --- > This is an automatically generated e-mail.

Re: Review Request 38378: Minor cleanup in perf_tests.cpp.

2015-09-14 Thread Cong Wang
tps://reviews.apache.org/r/38378/#comment155703> You are changing the semantic, the original test case is to verify a list of events which contains one invalid event, you changed to to verify just one invalid event. They are different. The original one makes more sense to me. - Cong Wang On Se

Re: Review Request 37540: Add perf event API

2015-09-15 Thread Cong Wang
only reads it when user calls read(). - Cong --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/37540/#review98924 --- On Sept. 4, 2

Re: Review Request 38233: os: add swap information to memory().

2015-09-11 Thread Cong Wang
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/38233/#review98629 --- Ship it! Ship It! - Cong Wang On Sept. 10, 2015, 5:57 p.m., Chi

Re: Review Request 38172: Stout: Simplified hashset implementation.

2015-09-15 Thread Cong Wang
17) <https://reviews.apache.org/r/38172/#comment155890> Why do you need cmath header here? - Cong Wang On Sept. 8, 2015, 9:52 p.m., Joris Van Remoortere wrote: > > --- > This is an automatically generated e-mail.

Re: Review Request 37540: Add perf event API

2015-09-29 Thread Cong Wang
the same result when they read from the same perf event handler. - Cong --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/37540/#review100656 ---

Re: Review Request 37540: Add perf event API

2015-09-29 Thread Cong Wang
) - src/linux/perf.hpp d10968ca670eb516bae08385fd0ddde8e8ad83b5 src/linux/perf.cpp f7035ddb2507a7646d88dd517d048018f695448a src/tests/containerizer/perf_tests.cpp ed5212ee31b8aa47339b8b8fab184bbdf85be82a Diff: https://reviews.apache.org/r/37540/diff/ Testing --- make check Thanks, Cong

Re: Review Request 38074: Calculate schedule latency with trace events

2015-09-29 Thread Cong Wang
f76f0f6a0c7a01c7f061a225d7f6ef52be0ee7b5 src/slave/flags.cpp 029aa1eb00e1aa3e92c2155925022c17fd905862 Diff: https://reviews.apache.org/r/38074/diff/ Testing --- manual tests Thanks, Cong Wang

Re: Review Request 37541: Add trace event API

2015-09-29 Thread Cong Wang
ed5212ee31b8aa47339b8b8fab184bbdf85be82a Diff: https://reviews.apache.org/r/37541/diff/ Testing --- make check Thanks, Cong Wang

Re: Review Request 38817: Added a SIGPIPE handler for the libprocess tests.

2015-09-28 Thread Cong Wang
tps://reviews.apache.org/r/38817/#comment158133> Here you probably want a signal string (strsignal ()) instead of a raw value. - Cong Wang On Sept. 28, 2015, 7:29 p.m., Ben Mahler wrote: > > --- > This is an automatically g

Re: Review Request 36180: Avoid multi writers write to same file in PortMappingIsolatorTests.

2015-10-05 Thread Cong Wang
anything to that file when nc receives nothing. - Cong Wang On Oct. 3, 2015, 4:52 p.m., haosdent huang wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache

Re: Review Request 36180: Avoid multi writers write to same file in PortMappingIsolatorTests.

2015-10-05 Thread Cong Wang
> On Oct. 6, 2015, 12:01 a.m., Cong Wang wrote: > > Or simply use 'tee -a' instead? Since it should not append anything to that > > file when nc receives nothing. > > haosdent huang wrote: > tee -a aslo OK, but seems use different files would more clear? I have

Re: Review Request 39013: RegistryClient refactor: Fixed comments style.

2015-10-05 Thread Cong Wang
-3523 to catch this coding style issue in future? - Cong Wang On Oct. 6, 2015, 2:30 a.m., Jojy Varghese wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache

Re: Review Request 36180: Avoid multi writers write to same file in PortMappingIsolatorTests.

2015-10-05 Thread Cong Wang
tps://reviews.apache.org/r/36180/#comment159024> traffic_invalid_via_loopback is not good either, because there is no traffic from an invalid port.... - Cong Wang On Oct. 6, 2015, 3:13 a.m., haosdent huang

Re: Review Request 38639: Add os::clone function in stout.

2015-09-23 Thread Cong Wang
> On Sept. 22, 2015, 6:55 p.m., Cong Wang wrote: > > 3rdparty/libprocess/3rdparty/stout/include/stout/os/linux.hpp, line 52 > > <https://reviews.apache.org/r/38639/diff/1/?file=1081455#file1081455line52> > > > > s/namespaces/flags/, because SIG* can

Re: Review Request 38569: Made slave's work_dir a shared mount in LinuxFilesystemIsolator.

2015-09-22 Thread Cong Wang
/reviews.apache.org/r/38569/#comment156941> This one looks unnecessary. - Cong Wang On Sept. 21, 2015, 9:23 p.m., Jie Yu wrote: > > --- > This is an automatically generated e-mail. To reply, visit: >

Re: Review Request 38535: Fix ExamplesTest errors when user is root.

2015-09-22 Thread Cong Wang
tps://reviews.apache.org/r/38535/#comment156949> You forgot to remove this line? There are two more places calling clone() in a similar way: src/tests/containerizer/launch_tests.cpp, src/tests/containerizer/ns_tests.cpp. Not sure if you need to fix them too. - Cong Wang On Sept. 20, 2015

Re: Review Request 38625: Keep clone syscall consistent.

2015-09-22 Thread Cong Wang
'? - Cong Wang On Sept. 22, 2015, 2:01 p.m., haosdent huang wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache

Re: Review Request 37967: Added Non-Freezeer Task Killer. 36620

2015-09-22 Thread Cong Wang
g/r/37967/#comment157179> s/verified/error/, otherwise reads odd. src/linux/cgroups.cpp (line 1631) <https://reviews.apache.org/r/37967/#comment157164> s/chain/statuses/ - Cong Wang On Sept. 22, 2015, 7:43 p.m.,

Re: Review Request 38639: Add os::clone function in stout.

2015-09-22 Thread Cong Wang
> On Sept. 22, 2015, 6:55 p.m., Cong Wang wrote: > > 3rdparty/libprocess/3rdparty/stout/include/stout/os/linux.hpp, line 52 > > <https://reviews.apache.org/r/38639/diff/1/?file=1081455#file1081455line52> > > > > s/namespaces/flags/, because SIG* can

  1   2   3   >