Hi
I was running Solr tests now and thought they hung, but eventually they
continued and I noticed that SimplePostToolTest took 280s to complete. I
tried from eclipse, and it took 1s. Tried from Ant again, 276s. I compared
(briefly) the outputs of the test from eclipse and Ant, and they look
simil
Can you check if you're using the same virtual machine from Eclipse
and from command line? Maybe it's a different
vendor/architecture/version?
I will look into this if you can't figure it out before me. Just give
me the full ANT line you were using.
Dawid
On Fri, Sep 13, 2013 at 6:15 PM, Shai Er
It passes when run from Eclipse? It seems crazy that it can take 280
sec from ant but 1 sec from Eclipse. Maybe it's not actually running,
when running from Eclipse?
+1 for @Nightly it we can't get to the bottom of it ...
Mike McCandless
http://blog.mikemccandless.com
On Thu, Sep 12, 2013 at
Yes it happens with the same seed. And it happened both times I ran it:
from Ant it takes ~280s, from eclipse ~1s.
Both times the test seems successful, and the outputs looks the same, so I
guess it means it does the same thing when run from eclipse and ant.
Am I the only one that experiences thi
Is it with the same seed?
Dawid
On Fri, Sep 13, 2013 at 1:04 PM, Michael McCandless
wrote:
> It passes when run from Eclipse? It seems crazy that it can take 280
> sec from ant but 1 sec from Eclipse. Maybe it's not actually running,
> when running from Eclipse?
>
> +1 for @Nightly it we can't
Sounds more like a timeout. I mean, posting documents should be reasonably
fast.
-- Jack Krupansky
-Original Message-
From: Michael McCandless
Sent: Friday, September 13, 2013 7:04 AM
To: Lucene/Solr dev
Subject: Re: SimplePostToolTest very slow
It passes when run from Eclipse? It
: Am I the only one that experiences this slowness?
I can reproduce the speeds you are seeing from ant (no IDE to test from on
my end)
It looks like some sort of delay in init ... every test has ~25s delay,
and changing the SimplePostTool instances to static and switching the
@Before to @Befo
: and changing the SimplePostTool instances to static and switching the
: @Before to @BeforeClass causes the whole tests runtime to drop down to 50
: seconds for me.
i couldn't leave that change in because it introduced failures depending
on test ordering (i thought those SimplePostTool object
Does this test actually try to connect to those URLs? If not then a
fake:// URL handler would be a very elegant solution not reaching to
the DNS subsystem at all? Not that I want to write it -- I remember it
was kind of nightmarish :)
Dawid
On Fri, Sep 13, 2013 at 8:48 PM, Chris Hostetter
wrote:
With the patch on SOLR-5241, the test runs for 1.8-3.2s, so this seems to
solve the problem.
I'm running w/ IBM J9 1.7.0, but it also happens w/ Oracle 1.7.0 (seems to
be even slower!), in both Ant and Eclipse.
You can reproduce by running "ant test -Dtestcase=SimplePostToolTest" from
/solr/core.
Loggers
On Fri, Sep 13, 2013 at 4:22 PM, Dawid Weiss
wrote:
> I still don't see why you'd get different timings between Eclipse and
> Ant if you're running with the same VM -- it should be pretty
> consistent (either it caches dns lookups or it doesn't).
>
> D.
>
>
> On Fri, Sep 13, 2013 at 10:01
I still don't see why you'd get different timings between Eclipse and
Ant if you're running with the same VM -- it should be pretty
consistent (either it caches dns lookups or it doesn't).
D.
On Fri, Sep 13, 2013 at 10:01 PM, Shai Erera wrote:
> With the patch on SOLR-5241, the test runs for 1.
They should actually be faster in Ant than they are in Eclipse
(because Eclipse displays logs in the console and Ant flushes them to
a file)?
Dawid
On Fri, Sep 13, 2013 at 10:29 PM, Robert Muir wrote:
> Loggers
>
> On Fri, Sep 13, 2013 at 4:22 PM, Dawid Weiss
> wrote:
>> I still don't see why y
: I still don't see why you'd get different timings between Eclipse and
: Ant if you're running with the same VM -- it should be pretty
: consistent (either it caches dns lookups or it doesn't).
Maybe Eclipse mucks with networkaddress.cache.ttl &
networkaddress.cache.negative.ttl ?
If you want
: If you want to experiment, a really trivial test is below -- on my system,
: there is a ~5 second gap between each pair of "INIT" and "H1" timestamps,
: but no other odd gaps in subsequent timestamps -- suggesting no caching of
: DNS per hostname, but that the URL class doesn't "re-lookup" on
>
> I still don't see why you'd get different timings between Eclipse and
> Ant if you're running with the same VM -- it should be pretty
> consistent (either it caches dns lookups or it doesn't).
>
I agree, it's suspicious. I searched for URL performance differences
between eclipse and Ant and hi
OK, I think I've made some progress -- configuration issue, but not
proxies, it seems to be our security manager/policy.
I printed system properties and env in setUp and ran only a single test
(testIsOn) to compare the output. I didn't notice any proxy settings, but I
did notice that from Ant we'r
ail: u...@thetaphi.de
From: Shai Erera [mailto:ser...@gmail.com]
Sent: Saturday, September 14, 2013 6:32 AM
To: dev@lucene.apache.org
Subject: Re: SimplePostToolTest very slow
OK, I think I've made some progress -- configuration issue, but not proxies, it
seems to be our security manager
: the main problem with any security manager is: To check if a connection
: is allowed, it has to resolve DNS and look the IP up in the policy.
Can we update the secuity policy to fail fast anytime a DNS lookup
happens? even if it happens implicitly in situations like this
(URL.hashCode) so we
Solr tests will all completely fail in that case then: just like they
do when i run them on my laptop with internet disconnected.
thats because it looks up its own hostname: which involves
reverse/forward dns lookups.
On Mon, Sep 16, 2013 at 1:07 PM, Chris Hostetter
wrote:
>
> : the main problem
20 matches
Mail list logo