How long should make check normally take to run for a serial build? After 'make check' was hanging for me at testphdf5 when building HDF5 with parallel support, make check now seems to be hanging when testing btree2. It's been at the same spot for 30 minutes or more, but this time, I see the process (lt-btree2) using CPU and memory in top, just not much of it. Is this normal? I would expect file I/O tests to only take a few minutes, not 30+ minutes.
--
Prentice
*On 07/25/2016 02:59 PM, Prentice Bisbal wrote:*
How long should make check normally take to run for a serial
build? After 'make check' was hanging for me at testphdf5 when
building HDF5 with parallel support, make check now seems to be
hanging when testing btree2. It's been at the same spot for 30
minutes or more, but this time, I see the process (lt-btree2)
using CPU and memory in top, just not much of it. Is this normal?
I would expect file I/O tests to only take a few minutes, not 30+
minutes.
Prentice -- sorry I can't solve the problem of the make check hanging, but I use the following setup which might be useful to you too:
export HDF5_ALARM_SECONDS=180 # Reduce timeout for bad case.
export MPIEXEC_TIMEOUT=360 # Give HDF the chance to
fail first.
set +e # My script uses bash -ex,
but we don't want to exit-on-error here.
make -i -k check # Make sure everything runs.
Use -i so it continues past erroneous cases.
set -e
So the bad cases abort after 3-4 minutes (depending on whether the HDF5 test-driver times out, or the MPI daemon).
But the +e and -i -k flags force all the tests to get run, so at least I get a complete picture of what worked and what didn't.
I'm running into problems with applications that use HDF5/NetCDF/PIO hanging or failing and am wondering if this is an issue with my file system as well.
Thanks,
Carl Ponder
-----------------------------------------------------------------------------------
This email message is for the sole use of the intended recipient(s) and may contain
confidential information. Any unauthorized review, use, disclosure or distribution
is prohibited. If you are not the intended recipient, please contact the sender by
reply email and destroy all copies of the original message.
-----------------------------------------------------------------------------------
30+mins *does* seem a bit long.
Typically, a ‘make check’ for me completes *all* tests in a total of about 10-15 mins at the most.
Where (e.g. on which kind of filesystem) are you trying to run ‘make check’?
For parallel tests, you *should* be using a parallel fileystem. I frequently forget about that.
If the parallel filesystem is *extremely* busy and/or not setup to deal with small I/Os and/or files (as the tests generate), I could see there being *some* additional delay in completion.
The btree test I think is testing HDF5’s internal btree data structures for group objects. So, if its doing any I/O I think its a bunch of tiny HDF5 metadata stuff.
Can you kill -9 that one process and see if it proceeds to completion on remaining tests?
Mark
From: Hdf-forum <hdf-forum-bounces@lists.hdfgroup.org<mailto:hdf-forum-bounces@lists.hdfgroup.org>> on behalf of Prentice Bisbal <pbisbal@pppl.gov<mailto:pbisbal@pppl.gov>>
Reply-To: HDF Users Discussion List <hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>>
Date: Monday, July 25, 2016 at 12:59 PM
To: "hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>" <hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>>
Subject: [Hdf-forum] How long should 'make check' take?
How long should make check normally take to run for a serial build?
After 'make check' was hanging for me at testphdf5 when building HDF5
with parallel support, make check now seems to be hanging when testing
btree2. It's been at the same spot for 30 minutes or more, but this
time, I see the process (lt-btree2) using CPU and memory in top, just
not much of it. Is this normal? I would expect file I/O tests to only
take a few minutes, not 30+ minutes.
--
Prentice
_______________________________________________
Hdf-forum is for HDF software users discussion.
Hdf-forum@lists.hdfgroup.org<mailto:Hdf-forum@lists.hdfgroup.org>
http://lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org
Twitter: https://twitter.com/hdf5
Where (e.g. on which kind of filesystem) are you trying to run �make check�?
Can you kill -9 that one process and see if it proceeds to completion on remaining tests?
Since I'm working on a serial build, right now, I'm using scratch disk local to the node I'm doing my builds on, mounted as '/local' When I do a control-C, the whole 'make check' process appears to end, it doesn't go on to the next test.
I'm now doing a build with minimal options specified in the configure step to see if maybe combinations of options was causing the problem.
For parallel tests, you *should* be using a parallel fileystem. I frequently forget about that.
Do you mean a shared filesystem, like NFS, or a true parallel filesystem like Lustre or GPFS? After realizing my stupidity of trying to do the parallel make check on my local disk, I did move it to an NFS filesystem.
Prentice
On 07/25/2016 04:07 PM, Miller, Mark C. wrote:
30+mins *does* seem a bit long.
Typically, a �make check� for me completes *all* tests in a total of about 10-15 mins at the most.
Where (e.g. on which kind of filesystem) are you trying to run �make check�?
For parallel tests, you *should* be using a parallel fileystem. I frequently forget about that.
If the parallel filesystem is *extremely* busy and/or not setup to deal with small I/Os and/or files (as the tests generate), I could see there being *some* additional delay in completion.
The btree test I think is testing HDF5�s internal btree data structures for group objects. So, if its doing any I/O I think its a bunch of tiny HDF5 metadata stuff.
Can you kill -9 that one process and see if it proceeds to completion on remaining tests?
Mark
From: Hdf-forum <hdf-forum-bounces@lists.hdfgroup.org <mailto:hdf-forum-bounces@lists.hdfgroup.org>> on behalf of Prentice Bisbal <pbisbal@pppl.gov <mailto:pbisbal@pppl.gov>>
Reply-To: HDF Users Discussion List <hdf-forum@lists.hdfgroup.org <mailto:hdf-forum@lists.hdfgroup.org>>
Date: Monday, July 25, 2016 at 12:59 PM
To: "hdf-forum@lists.hdfgroup.org <mailto:hdf-forum@lists.hdfgroup.org>" <hdf-forum@lists.hdfgroup.org <mailto:hdf-forum@lists.hdfgroup.org>>
Subject: [Hdf-forum] How long should 'make check' take?How long should make check normally take to run for a serial build?
After 'make check' was hanging for me at testphdf5 when building HDF5
with parallel support, make check now seems to be hanging when
testing
btree2. It's been at the same spot for 30 minutes or more, but this
time, I see the process (lt-btree2) using CPU and memory in top, just
not much of it. Is this normal? I would expect file I/O tests to only
take a few minutes, not 30+ minutes.-- Prentice
_______________________________________________
Hdf-forum is for HDF software users discussion.
Hdf-forum@lists.hdfgroup.org <mailto:Hdf-forum@lists.hdfgroup.org>
http://lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org
Twitter: https://twitter.com/hdf5_______________________________________________
Hdf-forum is for HDF software users discussion.
Hdf-forum@lists.hdfgroup.org
http://lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org
Twitter: https://twitter.com/hdf5
Yeah, I meant ‘kill -9’ not ctrl-C. I would expect ctrl-C to just abort the current command (e.g. ‘make check’). By killing the running process (lt-btree) alone, make check would just interpret that as some kind of failure and then move on to the next test.
Yes, you need a “true” parallel filesystem for any parallel tests. NFS will NOT work.
Running on /local makes a lot of sense other than for parallel. Whenever I run on NFS, things take a lot longer due to all the contention for that shared resource.
Mark
From: Hdf-forum <hdf-forum-bounces@lists.hdfgroup.org<mailto:hdf-forum-bounces@lists.hdfgroup.org>> on behalf of Prentice Bisbal <pbisbal@pppl.gov<mailto:pbisbal@pppl.gov>>
Reply-To: HDF Users Discussion List <hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>>
Date: Monday, July 25, 2016 at 1:22 PM
To: "hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>" <hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>>
Subject: Re: [Hdf-forum] How long should 'make check' take?
Where (e.g. on which kind of filesystem) are you trying to run ‘make check’?
Can you kill -9 that one process and see if it proceeds to completion on remaining tests?
Since I'm working on a serial build, right now, I'm using scratch disk local to the node I'm doing my builds on, mounted as '/local' When I do a control-C, the whole 'make check' process appears to end, it doesn't go on to the next test.
I'm now doing a build with minimal options specified in the configure step to see if maybe combinations of options was causing the problem.
For parallel tests, you *should* be using a parallel fileystem. I frequently forget about that.
Do you mean a shared filesystem, like NFS, or a true parallel filesystem like Lustre or GPFS? After realizing my stupidity of trying to do the parallel make check on my local disk, I did move it to an NFS filesystem.
Prentice
On 07/25/2016 04:07 PM, Miller, Mark C. wrote:
30+mins *does* seem a bit long.
Typically, a ‘make check’ for me completes *all* tests in a total of about 10-15 mins at the most.
Where (e.g. on which kind of filesystem) are you trying to run ‘make check’?
For parallel tests, you *should* be using a parallel fileystem. I frequently forget about that.
If the parallel filesystem is *extremely* busy and/or not setup to deal with small I/Os and/or files (as the tests generate), I could see there being *some* additional delay in completion.
The btree test I think is testing HDF5’s internal btree data structures for group objects. So, if its doing any I/O I think its a bunch of tiny HDF5 metadata stuff.
Can you kill -9 that one process and see if it proceeds to completion on remaining tests?
Mark
From: Hdf-forum <hdf-forum-bounces@lists.hdfgroup.org<mailto:hdf-forum-bounces@lists.hdfgroup.org>> on behalf of Prentice Bisbal <pbisbal@pppl.gov<mailto:pbisbal@pppl.gov>>
Reply-To: HDF Users Discussion List <hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>>
Date: Monday, July 25, 2016 at 12:59 PM
To: "hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>" <hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>>
Subject: [Hdf-forum] How long should 'make check' take?
How long should make check normally take to run for a serial build?
After 'make check' was hanging for me at testphdf5 when building HDF5
with parallel support, make check now seems to be hanging when testing
btree2. It's been at the same spot for 30 minutes or more, but this
time, I see the process (lt-btree2) using CPU and memory in top, just
not much of it. Is this normal? I would expect file I/O tests to only
take a few minutes, not 30+ minutes.
--
Prentice
_______________________________________________
Hdf-forum is for HDF software users discussion.
Hdf-forum@lists.hdfgroup.org<mailto:Hdf-forum@lists.hdfgroup.org>
http://lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org
Twitter: https://twitter.com/hdf5
_______________________________________________
Hdf-forum is for HDF software users discussion.
Hdf-forum@lists.hdfgroup.org<mailto:Hdf-forum@lists.hdfgroup.org>http://lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org
Twitter: https://twitter.com/hdf5
You did say kill -9, which my brain wrongly interpreted as Control-C. Sorry. I configured HDF5 with minimal options:
./configure --prefix=/usr/pppl/gcc/5.4-pkgs/hdf5-serial-1.10.0-patch1 CC=gcc CXX=g++ FC=gfortran 2>&1 | tee configure.log
When I got to the make check stage, I noted the time testing btree started. After 20 minutes I killed it (didn't even need the -9), and it spit out the output shown below. . As you can see at the bottom, the make check then exited with an error.
Any suggestions on how to proceed? Should I run lt-bree2 manually and run a stack trace on it or do anything like that?
Part of me is ready to install w/o the make check completing successfully, but I have a feeling that will just come back to haunt me in the near future.
Testing btree2
Testing without reopening B-tree:
Testing B-tree creation PASSED
Testing B-tree iteration: empty B-tree PASSED
Testing B-tree insert: first record PASSED
Testing B-tree insert: several records PASSED
Testing B-tree insert: split root PASSED
Testing B-tree insert: redistribute 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree insert: redistribute 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree insert: split side leaf into 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree insert: split side leaf into 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree insert: redistribute 3 leaves in level 1 B-tree PASSED
Testing B-tree insert: split middle leaf into 2 leaves in level 1 B-tree PASSED
Testing B-tree insert: make level 2 B-tree PASSED
Testing B-tree insert: redistrib right-most leaf in level 2 B-tree PASSED
Testing B-tree insert: redistrib left-most leaf in level 2 B-tree PASSED
Testing B-tree insert: redistrib middle leaf in level 2 B-tree PASSED
Testing B-tree insert: split right-most leaf in level 2 B-tree PASSED
Testing B-tree insert: split left-most leaf in level 2 B-tree PASSED
Testing B-tree insert: split middle leaf in level 2 B-tree PASSED
Testing B-tree insert: redist. 2 internal (r->l) in level 2 B-tree PASSED
Testing B-tree insert: redist. 2 internal (l->r) in level 2 B-tree PASSED
Testing B-tree insert: split side internal node to 2 in level 2 B-tree (r->l) PASSED
Testing B-tree insert: split side internal node to 2 in level 2 B-tree (l->2) PASSED
Testing B-tree insert: redistrib 3 internals in level 2 B-tree PASSED
Testing B-tree insert: split 3 internals to 4 in level 2 B-tree PASSED
Testing B-tree insert: create random level 4 B-tree PASSED
Testing B-tree insert: attempt duplicate record in level 4 B-tree PASSED
Testing B-tree update: inserting first record in empty B-tree PASSED
Testing B-tree update: update only record in B-tree PASSED
Testing B-tree update: insert several records PASSED
Testing B-tree update: update several records PASSED
Testing B-tree update: split root PASSED
Testing B-tree update: redistribute 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree update: redistribute 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree update: split side leaf into 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree update: split side leaf into 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree update: redistribute 3 leaves in level 1 B-tree PASSED
Testing B-tree update: split middle leaf into 2 leaves in level 1 B-tree PASSED
Testing B-tree update: make level 2 B-tree (l->r) PASSED
Testing B-tree update: make level 2 B-tree (r->l) PASSED
Testing B-tree update: make level 2 B-tree (l+r->middle) PASSED
Testing B-tree update: create random level 4 B-tree PASSED
Testing B-tree update: update record in level 4 B-tree PASSED
Testing B-tree remove: record from empty B-tree PASSED
Testing B-tree remove: non-existant record from 1 record B-tree PASSED
Testing B-tree remove: existant record from 1 record B-tree PASSED
Testing B-tree remove: adding records to B-tree after removal PASSED
Testing B-tree remove: non-existant record from level-0 B-tree PASSED
Testing B-tree remove: mult. existant records from level-0 B-tree PASSED
Testing B-tree remove: non-existant record from level-1 B-tree PASSED
Testing B-tree remove: record from right leaf of level-1 B-tree PASSED
Testing B-tree remove: record from left leaf of level-1 B-tree PASSED
Testing B-tree remove: record from middle leaf of level-1 B-tree PASSED
Testing B-tree remove: redistribute 2 leaves in level-1 B-tree (r->l) PASSED
Testing B-tree remove: redistribute 2 leaves in level-1 B-tree (l->r) PASSED
Testing B-tree remove: redistribute 3 leaves in level-1 B-tree PASSED
Testing B-tree remove: merge 2 leaves to 1 in level-1 B-tree (r->l) PASSED
Testing B-tree remove: merge 2 leaves to 1 in level-1 B-tree (l->r) PASSED
Testing B-tree remove: merge 3 leaves to 2 in level-1 B-tree PASSED
Testing B-tree remove: promote from right leaf of level-1 B-tree PASSED
Testing B-tree remove: promote from left leaf of level-1 B-tree PASSED
Testing B-tree remove: promote from middle leaf of level-1 B-tree PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/2 node redistrib PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/3 node redistrib PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/2->1 merge PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/3->2 merge PASSED
Testing B-tree remove: collapse level-1 B-tree back to level-0 PASSED
Testing B-tree remove: promote from right internal of level-2 B-tree PASSED
Testing B-tree remove: promote from left internal of level-2 B-tree PASSED
Testing B-tree remove: promote from middle internal of level-2 B-tree PASSED
Testing B-tree remove: promote record from root of level-2 B-tree PASSED
Testing B-tree remove: promote from right internal of level-2 B-tree w/redistrib PASSED
Testing B-tree remove: promote from left internal of level-2 B-tree w/redistrib PASSED
Testing B-tree remove: promote from right internal of level-2 B-tree w/merge PASSED
Testing B-tree remove: promote from middle internal of level-2 B-tree w/merge PASSED
Testing B-tree remove: merge 2 internal nodes to 1 in level-2 B-tree (l->r) PASSED
Testing B-tree remove: merge 2 internal nodes to 1 in level-2 B-tree (r->l) PASSED
Testing B-tree remove: merge 3 internal nodes to 2 in level-2 B-tree PASSED
Testing B-tree remove: collapse level-2 B-tree back to level-1 (r->l) PASSED
Testing B-tree remove: create random level 4 B-tree and delete all records in random order PASSED
Testing B-tree remove: create random level 4 B-tree and delete all records by index, in random order PASSED
Testing B-tree remove: create random level 4 B-tree and delete all records by index, in increasing order PASSED
Testing B-tree remove: create random level 4 B-tree and delete all records by index, in decreasing order PASSED
Testing B-tree find: nearest neighbor less than a value PASSED
Testing B-tree find: nearest neighbor greater than a value PASSED
Testing B-tree delete: delete empty B-tree PASSED
Testing B-tree delete: delete level-0 B-tree PASSED
Testing B-tree delete: delete level-1 B-tree PASSED
Testing B-tree delete: delete level-2 B-tree PASSED
Testing B-tree modify: attempt to modify non-existant record PASSED
Testing B-tree modify: modify record in leaf node PASSED
Testing B-tree modify: modify record in internal node PASSED
Testing B-tree modify: modify record in root node PASSED
Testing with reopening B-tree:
Testing B-tree creation PASSED
Testing B-tree iteration: empty B-tree PASSED
Testing B-tree insert: first record PASSED
Testing B-tree insert: several records PASSED
Testing B-tree insert: split root PASSED
Testing B-tree insert: redistribute 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree insert: redistribute 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree insert: split side leaf into 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree insert: split side leaf into 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree insert: redistribute 3 leaves in level 1 B-tree PASSED
Testing B-tree insert: split middle leaf into 2 leaves in level 1 B-tree PASSED
Testing B-tree insert: make level 2 B-tree PASSED
Testing B-tree insert: redistrib right-most leaf in level 2 B-tree PASSED
Testing B-tree insert: redistrib left-most leaf in level 2 B-tree PASSED
Testing B-tree insert: redistrib middle leaf in level 2 B-tree PASSED
Testing B-tree insert: split right-most leaf in level 2 B-tree PASSED
Testing B-tree insert: split left-most leaf in level 2 B-tree PASSED
Testing B-tree insert: split middle leaf in level 2 B-tree PASSED
Testing B-tree insert: redist. 2 internal (r->l) in level 2 B-tree PASSED
Testing B-tree insert: redist. 2 internal (l->r) in level 2 B-tree PASSED
Testing B-tree insert: split side internal node to 2 in level 2 B-tree (r->l) PASSED
Testing B-tree insert: split side internal node to 2 in level 2 B-tree (l->2) PASSED
Testing B-tree insert: redistrib 3 internals in level 2 B-tree PASSED
Testing B-tree insert: split 3 internals to 4 in level 2 B-tree PASSED
Testing B-tree insert: create random level 4 B-tree PASSED
Testing B-tree insert: attempt duplicate record in level 4 B-tree PASSED
Testing B-tree update: inserting first record in empty B-tree PASSED
Testing B-tree update: update only record in B-tree PASSED
Testing B-tree update: insert several records PASSED
Testing B-tree update: update several records PASSED
Testing B-tree update: split root PASSED
Testing B-tree update: redistribute 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree update: redistribute 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree update: split side leaf into 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree update: split side leaf into 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree update: redistribute 3 leaves in level 1 B-tree PASSED
Testing B-tree update: split middle leaf into 2 leaves in level 1 B-tree PASSED
Testing B-tree update: make level 2 B-tree (l->r) PASSED
Testing B-tree update: make level 2 B-tree (r->l) PASSED
Testing B-tree update: make level 2 B-tree (l+r->middle) PASSED
Testing B-tree update: create random level 4 B-tree PASSED
Testing B-tree update: update record in level 4 B-tree PASSED
Testing B-tree remove: record from empty B-tree PASSED
Testing B-tree remove: non-existant record from 1 record B-tree PASSED
Testing B-tree remove: existant record from 1 record B-tree PASSED
Testing B-tree remove: adding records to B-tree after removal PASSED
Testing B-tree remove: non-existant record from level-0 B-tree PASSED
Testing B-tree remove: mult. existant records from level-0 B-tree PASSED
Testing B-tree remove: non-existant record from level-1 B-tree PASSED
Testing B-tree remove: record from right leaf of level-1 B-tree PASSED
Testing B-tree remove: record from left leaf of level-1 B-tree PASSED
Testing B-tree remove: record from middle leaf of level-1 B-tree PASSED
Testing B-tree remove: redistribute 2 leaves in level-1 B-tree (r->l) PASSED
Testing B-tree remove: redistribute 2 leaves in level-1 B-tree (l->r) PASSED
Testing B-tree remove: redistribute 3 leaves in level-1 B-tree PASSED
Testing B-tree remove: merge 2 leaves to 1 in level-1 B-tree (r->l) PASSED
Testing B-tree remove: merge 2 leaves to 1 in level-1 B-tree (l->r) PASSED
Testing B-tree remove: merge 3 leaves to 2 in level-1 B-tree PASSED
Testing B-tree remove: promote from right leaf of level-1 B-tree PASSED
Testing B-tree remove: promote from left leaf of level-1 B-tree PASSED
Testing B-tree remove: promote from middle leaf of level-1 B-tree PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/2 node redistrib PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/3 node redistrib PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/2->1 merge PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/3->2 merge PASSED
Testing B-tree remove: collapse level-1 B-tree back to level-0 PASSED
Testing B-tree remove: promote from right internal of level-2 B-tree PASSED
Testing B-tree remove: promote from left internal of level-2 B-tree PASSED
Testing B-tree remove: promote from middle internal of level-2 B-tree PASSED
Testing B-tree remove: promote record from root of level-2 B-tree PASSED
Testing B-tree remove: promote from right internal of level-2 B-tree w/redistrib PASSED
Testing B-tree remove: promote from left internal of level-2 B-tree w/redistrib PASSED
Testing B-tree remove: promote from right internal of level-2 B-tree w/merge PASSED
Testing B-tree remove: promote from middle internal of level-2 B-tree w/merge PASSED
Testing B-tree remove: merge 2 internal nodes to 1 in level-2 B-tree (l->r) PASSED
Testing B-tree remove: merge 2 internal nodes to 1 in level-2 B-tree (r->l) PASSED
Testing B-tree remove: merge 3 internal nodes to 2 in level-2 B-tree PASSED
Testing B-tree remove: collapse level-2 B-tree back to level-1 (r->l) PASSED
Command terminated by signal 15
150.13user 39.62system 22:25.95elapsed 14%CPU (0avgtext+0avgdata 494240maxresident)k
0inputs+0outputs (0major+54831minor)pagefaults 0swaps
make[4]: *** [btree2.chkexe_] Error 1
make[4]: Leaving directory `/local/pbisbal/gcc-5.4.0/hdf5-1.10.0-patch1/test'
make[3]: *** [build-check-s] Error 2
make[3]: Leaving directory `/local/pbisbal/gcc-5.4.0/hdf5-1.10.0-patch1/test'
make[2]: *** [test] Error 2
make[2]: Leaving directory `/local/pbisbal/gcc-5.4.0/hdf5-1.10.0-patch1/test'
make[1]: *** [check-am] Error 2
make[1]: Leaving directory `/local/pbisbal/gcc-5.4.0/hdf5-1.10.0-patch1/test'
make: *** [check-recursive] Error 1
On 07/25/2016 04:31 PM, Miller, Mark C. wrote:
Yeah, I meant �kill -9� not ctrl-C. I would expect ctrl-C to just abort the current command (e.g. �make check�). By killing the running process (lt-btree) alone, make check would just interpret that as some kind of failure and then move on to the next test.
Yes, you need a �true� parallel filesystem for any parallel tests. NFS will NOT work.
Running on /local makes a lot of sense other than for parallel. Whenever I run on NFS, things take a lot longer due to all the contention for that shared resource.
Mark
From: Hdf-forum <hdf-forum-bounces@lists.hdfgroup.org <mailto:hdf-forum-bounces@lists.hdfgroup.org>> on behalf of Prentice Bisbal <pbisbal@pppl.gov <mailto:pbisbal@pppl.gov>>
Reply-To: HDF Users Discussion List <hdf-forum@lists.hdfgroup.org <mailto:hdf-forum@lists.hdfgroup.org>>
Date: Monday, July 25, 2016 at 1:22 PM
To: "hdf-forum@lists.hdfgroup.org <mailto:hdf-forum@lists.hdfgroup.org>" <hdf-forum@lists.hdfgroup.org <mailto:hdf-forum@lists.hdfgroup.org>>
Subject: Re: [Hdf-forum] How long should 'make check' take?Where (e.g. on which kind of filesystem) are you trying to run
�make check�?
Can you kill -9 that one process and see if it proceeds to
completion on remaining tests?Since I'm working on a serial build, right now, I'm using scratch
disk local to the node I'm doing my builds on, mounted as '/local'
When I do a control-C, the whole 'make check' process appears to
end, it doesn't go on to the next test.I'm now doing a build with minimal options specified in the
configure step to see if maybe combinations of options was causing
the problem.For parallel tests, you *should* be using a parallel fileystem. I
frequently forget about that.Do you mean a shared filesystem, like NFS, or a true parallel
filesystem like Lustre or GPFS? After realizing my stupidity of
trying to do the parallel make check on my local disk, I did move
it to an NFS filesystem.Prentice
On 07/25/2016 04:07 PM, Miller, Mark C. wrote:
30+mins *does* seem a bit long.
Typically, a �make check� for me completes *all* tests in a total
of about 10-15 mins at the most.Where (e.g. on which kind of filesystem) are you trying to run
�make check�?For parallel tests, you *should* be using a parallel fileystem. I
frequently forget about that.If the parallel filesystem is *extremely* busy and/or not setup
to deal with small I/Os and/or files (as the tests generate), I
could see there being *some* additional delay in completion.The btree test I think is testing HDF5�s internal btree data
structures for group objects. So, if its doing any I/O I think
its a bunch of tiny HDF5 metadata stuff.Can you kill -9 that one process and see if it proceeds to
completion on remaining tests?Mark
From: Hdf-forum <hdf-forum-bounces@lists.hdfgroup.org
<mailto:hdf-forum-bounces@lists.hdfgroup.org>> on behalf of
Prentice Bisbal <pbisbal@pppl.gov <mailto:pbisbal@pppl.gov>>
Reply-To: HDF Users Discussion List <hdf-forum@lists.hdfgroup.org
<mailto:hdf-forum@lists.hdfgroup.org>>
Date: Monday, July 25, 2016 at 12:59 PM
To: "hdf-forum@lists.hdfgroup.org
<mailto:hdf-forum@lists.hdfgroup.org>"
<hdf-forum@lists.hdfgroup.org <mailto:hdf-forum@lists.hdfgroup.org>>
Subject: [Hdf-forum] How long should 'make check' take?How long should make check normally take to run for a serial
build?
After 'make check' was hanging for me at testphdf5 when
building HDF5
with parallel support, make check now seems to be hanging
when testing
btree2. It's been at the same spot for 30 minutes or more,
but this
time, I see the process (lt-btree2) using CPU and memory in
top, just
not much of it. Is this normal? I would expect file I/O tests
to only
take a few minutes, not 30+ minutes.-- Prentice
_______________________________________________
Hdf-forum is for HDF software users discussion.
Hdf-forum@lists.hdfgroup.org
<mailto:Hdf-forum@lists.hdfgroup.org>
http://lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org
Twitter: https://twitter.com/hdf5_______________________________________________
Hdf-forum is for HDF software users discussion.
Hdf-forum@lists.hdfgroup.orghttp://lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org
Twitter:https://twitter.com/hdf5_______________________________________________
Hdf-forum is for HDF software users discussion.
Hdf-forum@lists.hdfgroup.org
http://lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org
Twitter: https://twitter.com/hdf5
Ordinarily, I’d agree. In fact, I don’t often run ‘make check’ on HDF5 due to the likelihood of tests themselves creating extra work (at least relative to the actual use cases for HDF5 callers I support).
This looks like a pretty vanilla build with GNU compilers. Are you using bleeding edge compiler versions?
If not, seems like it could be a bug in HDF5 testing or library. You might wanna ping The HDF Group help line directly about it.
I agree its an uneasy feeling to go ahead and install it with a known failure.
Mark
From: Hdf-forum <hdf-forum-bounces@lists.hdfgroup.org<mailto:hdf-forum-bounces@lists.hdfgroup.org>> on behalf of Prentice Bisbal <pbisbal@pppl.gov<mailto:pbisbal@pppl.gov>>
Reply-To: HDF Users Discussion List <hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>>
Date: Monday, July 25, 2016 at 2:05 PM
To: "hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>" <hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>>
Subject: Re: [Hdf-forum] How long should 'make check' take?
You did say kill -9, which my brain wrongly interpreted as Control-C. Sorry. I configured HDF5 with minimal options:
./configure --prefix=/usr/pppl/gcc/5.4-pkgs/hdf5-serial-1.10.0-patch1 CC=gcc CXX=g++ FC=gfortran 2>&1 | tee configure.log
When I got to the make check stage, I noted the time testing btree started. After 20 minutes I killed it (didn't even need the -9), and it spit out the output shown below. . As you can see at the bottom, the make check then exited with an error.
Any suggestions on how to proceed? Should I run lt-bree2 manually and run a stack trace on it or do anything like that?
Part of me is ready to install w/o the make check completing successfully, but I have a feeling that will just come back to haunt me in the near future.
Testing without reopening B-tree:
Testing B-tree creation PASSED
Testing B-tree iteration: empty B-tree PASSED
Testing B-tree insert: first record PASSED
Testing B-tree insert: several records PASSED
Testing B-tree insert: split root PASSED
Testing B-tree insert: redistribute 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree insert: redistribute 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree insert: split side leaf into 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree insert: split side leaf into 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree insert: redistribute 3 leaves in level 1 B-tree PASSED
Testing B-tree insert: split middle leaf into 2 leaves in level 1 B-tree PASSED
Testing B-tree insert: make level 2 B-tree PASSED
Testing B-tree insert: redistrib right-most leaf in level 2 B-tree PASSED
Testing B-tree insert: redistrib left-most leaf in level 2 B-tree PASSED
Testing B-tree insert: redistrib middle leaf in level 2 B-tree PASSED
Testing B-tree insert: split right-most leaf in level 2 B-tree PASSED
Testing B-tree insert: split left-most leaf in level 2 B-tree PASSED
Testing B-tree insert: split middle leaf in level 2 B-tree PASSED
Testing B-tree insert: redist. 2 internal (r->l) in level 2 B-tree PASSED
Testing B-tree insert: redist. 2 internal (l->r) in level 2 B-tree PASSED
Testing B-tree insert: split side internal node to 2 in level 2 B-tree (r->l) PASSED
Testing B-tree insert: split side internal node to 2 in level 2 B-tree (l->2) PASSED
Testing B-tree insert: redistrib 3 internals in level 2 B-tree PASSED
Testing B-tree insert: split 3 internals to 4 in level 2 B-tree PASSED
Testing B-tree insert: create random level 4 B-tree PASSED
Testing B-tree insert: attempt duplicate record in level 4 B-tree PASSED
Testing B-tree update: inserting first record in empty B-tree PASSED
Testing B-tree update: update only record in B-tree PASSED
Testing B-tree update: insert several records PASSED
Testing B-tree update: update several records PASSED
Testing B-tree update: split root PASSED
Testing B-tree update: redistribute 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree update: redistribute 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree update: split side leaf into 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree update: split side leaf into 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree update: redistribute 3 leaves in level 1 B-tree PASSED
Testing B-tree update: split middle leaf into 2 leaves in level 1 B-tree PASSED
Testing B-tree update: make level 2 B-tree (l->r) PASSED
Testing B-tree update: make level 2 B-tree (r->l) PASSED
Testing B-tree update: make level 2 B-tree (l+r->middle) PASSED
Testing B-tree update: create random level 4 B-tree PASSED
Testing B-tree update: update record in level 4 B-tree PASSED
Testing B-tree remove: record from empty B-tree PASSED
Testing B-tree remove: non-existant record from 1 record B-tree PASSED
Testing B-tree remove: existant record from 1 record B-tree PASSED
Testing B-tree remove: adding records to B-tree after removal PASSED
Testing B-tree remove: non-existant record from level-0 B-tree PASSED
Testing B-tree remove: mult. existant records from level-0 B-tree PASSED
Testing B-tree remove: non-existant record from level-1 B-tree PASSED
Testing B-tree remove: record from right leaf of level-1 B-tree PASSED
Testing B-tree remove: record from left leaf of level-1 B-tree PASSED
Testing B-tree remove: record from middle leaf of level-1 B-tree PASSED
Testing B-tree remove: redistribute 2 leaves in level-1 B-tree (r->l) PASSED
Testing B-tree remove: redistribute 2 leaves in level-1 B-tree (l->r) PASSED
Testing B-tree remove: redistribute 3 leaves in level-1 B-tree PASSED
Testing B-tree remove: merge 2 leaves to 1 in level-1 B-tree (r->l) PASSED
Testing B-tree remove: merge 2 leaves to 1 in level-1 B-tree (l->r) PASSED
Testing B-tree remove: merge 3 leaves to 2 in level-1 B-tree PASSED
Testing B-tree remove: promote from right leaf of level-1 B-tree PASSED
Testing B-tree remove: promote from left leaf of level-1 B-tree PASSED
Testing B-tree remove: promote from middle leaf of level-1 B-tree PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/2 node redistrib PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/3 node redistrib PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/2->1 merge PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/3->2 merge PASSED
Testing B-tree remove: collapse level-1 B-tree back to level-0 PASSED
Testing B-tree remove: promote from right internal of level-2 B-tree PASSED
Testing B-tree remove: promote from left internal of level-2 B-tree PASSED
Testing B-tree remove: promote from middle internal of level-2 B-tree PASSED
Testing B-tree remove: promote record from root of level-2 B-tree PASSED
Testing B-tree remove: promote from right internal of level-2 B-tree w/redistrib PASSED
Testing B-tree remove: promote from left internal of level-2 B-tree w/redistrib PASSED
Testing B-tree remove: promote from right internal of level-2 B-tree w/merge PASSED
Testing B-tree remove: promote from middle internal of level-2 B-tree w/merge PASSED
Testing B-tree remove: merge 2 internal nodes to 1 in level-2 B-tree (l->r) PASSED
Testing B-tree remove: merge 2 internal nodes to 1 in level-2 B-tree (r->l) PASSED
Testing B-tree remove: merge 3 internal nodes to 2 in level-2 B-tree PASSED
Testing B-tree remove: collapse level-2 B-tree back to level-1 (r->l) PASSED
Testing B-tree remove: create random level 4 B-tree and delete all records in random order PASSED
Testing B-tree remove: create random level 4 B-tree and delete all records by index, in random order PASSED
Testing B-tree remove: create random level 4 B-tree and delete all records by index, in increasing order PASSED
Testing B-tree remove: create random level 4 B-tree and delete all records by index, in decreasing order PASSED
Testing B-tree find: nearest neighbor less than a value PASSED
Testing B-tree find: nearest neighbor greater than a value PASSED
Testing B-tree delete: delete empty B-tree PASSED
Testing B-tree delete: delete level-0 B-tree PASSED
Testing B-tree delete: delete level-1 B-tree PASSED
Testing B-tree delete: delete level-2 B-tree PASSED
Testing B-tree modify: attempt to modify non-existant record PASSED
Testing B-tree modify: modify record in leaf node PASSED
Testing B-tree modify: modify record in internal node PASSED
Testing B-tree modify: modify record in root node PASSED
Testing with reopening B-tree:
Testing B-tree creation PASSED
Testing B-tree iteration: empty B-tree PASSED
Testing B-tree insert: first record PASSED
Testing B-tree insert: several records PASSED
Testing B-tree insert: split root PASSED
Testing B-tree insert: redistribute 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree insert: redistribute 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree insert: split side leaf into 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree insert: split side leaf into 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree insert: redistribute 3 leaves in level 1 B-tree PASSED
Testing B-tree insert: split middle leaf into 2 leaves in level 1 B-tree PASSED
Testing B-tree insert: make level 2 B-tree PASSED
Testing B-tree insert: redistrib right-most leaf in level 2 B-tree PASSED
Testing B-tree insert: redistrib left-most leaf in level 2 B-tree PASSED
Testing B-tree insert: redistrib middle leaf in level 2 B-tree PASSED
Testing B-tree insert: split right-most leaf in level 2 B-tree PASSED
Testing B-tree insert: split left-most leaf in level 2 B-tree PASSED
Testing B-tree insert: split middle leaf in level 2 B-tree PASSED
Testing B-tree insert: redist. 2 internal (r->l) in level 2 B-tree PASSED
Testing B-tree insert: redist. 2 internal (l->r) in level 2 B-tree PASSED
Testing B-tree insert: split side internal node to 2 in level 2 B-tree (r->l) PASSED
Testing B-tree insert: split side internal node to 2 in level 2 B-tree (l->2) PASSED
Testing B-tree insert: redistrib 3 internals in level 2 B-tree PASSED
Testing B-tree insert: split 3 internals to 4 in level 2 B-tree PASSED
Testing B-tree insert: create random level 4 B-tree PASSED
Testing B-tree insert: attempt duplicate record in level 4 B-tree PASSED
Testing B-tree update: inserting first record in empty B-tree PASSED
Testing B-tree update: update only record in B-tree PASSED
Testing B-tree update: insert several records PASSED
Testing B-tree update: update several records PASSED
Testing B-tree update: split root PASSED
Testing B-tree update: redistribute 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree update: redistribute 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree update: split side leaf into 2 leaves in level 1 B-tree (l->r) PASSED
Testing B-tree update: split side leaf into 2 leaves in level 1 B-tree (r->l) PASSED
Testing B-tree update: redistribute 3 leaves in level 1 B-tree PASSED
Testing B-tree update: split middle leaf into 2 leaves in level 1 B-tree PASSED
Testing B-tree update: make level 2 B-tree (l->r) PASSED
Testing B-tree update: make level 2 B-tree (r->l) PASSED
Testing B-tree update: make level 2 B-tree (l+r->middle) PASSED
Testing B-tree update: create random level 4 B-tree PASSED
Testing B-tree update: update record in level 4 B-tree PASSED
Testing B-tree remove: record from empty B-tree PASSED
Testing B-tree remove: non-existant record from 1 record B-tree PASSED
Testing B-tree remove: existant record from 1 record B-tree PASSED
Testing B-tree remove: adding records to B-tree after removal PASSED
Testing B-tree remove: non-existant record from level-0 B-tree PASSED
Testing B-tree remove: mult. existant records from level-0 B-tree PASSED
Testing B-tree remove: non-existant record from level-1 B-tree PASSED
Testing B-tree remove: record from right leaf of level-1 B-tree PASSED
Testing B-tree remove: record from left leaf of level-1 B-tree PASSED
Testing B-tree remove: record from middle leaf of level-1 B-tree PASSED
Testing B-tree remove: redistribute 2 leaves in level-1 B-tree (r->l) PASSED
Testing B-tree remove: redistribute 2 leaves in level-1 B-tree (l->r) PASSED
Testing B-tree remove: redistribute 3 leaves in level-1 B-tree PASSED
Testing B-tree remove: merge 2 leaves to 1 in level-1 B-tree (r->l) PASSED
Testing B-tree remove: merge 2 leaves to 1 in level-1 B-tree (l->r) PASSED
Testing B-tree remove: merge 3 leaves to 2 in level-1 B-tree PASSED
Testing B-tree remove: promote from right leaf of level-1 B-tree PASSED
Testing B-tree remove: promote from left leaf of level-1 B-tree PASSED
Testing B-tree remove: promote from middle leaf of level-1 B-tree PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/2 node redistrib PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/3 node redistrib PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/2->1 merge PASSED
Testing B-tree remove: promote from leaf of level-1 B-tree w/3->2 merge PASSED
Testing B-tree remove: collapse level-1 B-tree back to level-0 PASSED
Testing B-tree remove: promote from right internal of level-2 B-tree PASSED
Testing B-tree remove: promote from left internal of level-2 B-tree PASSED
Testing B-tree remove: promote from middle internal of level-2 B-tree PASSED
Testing B-tree remove: promote record from root of level-2 B-tree PASSED
Testing B-tree remove: promote from right internal of level-2 B-tree w/redistrib PASSED
Testing B-tree remove: promote from left internal of level-2 B-tree w/redistrib PASSED
Testing B-tree remove: promote from right internal of level-2 B-tree w/merge PASSED
Testing B-tree remove: promote from middle internal of level-2 B-tree w/merge PASSED
Testing B-tree remove: merge 2 internal nodes to 1 in level-2 B-tree (l->r) PASSED
Testing B-tree remove: merge 2 internal nodes to 1 in level-2 B-tree (r->l) PASSED
Testing B-tree remove: merge 3 internal nodes to 2 in level-2 B-tree PASSED
Testing B-tree remove: collapse level-2 B-tree back to level-1 (r->l) PASSED
Command terminated by signal 15
150.13user 39.62system 22:25.95elapsed 14%CPU (0avgtext+0avgdata 494240maxresident)k
0inputs+0outputs (0major+54831minor)pagefaults 0swaps
make[4]: *** [btree2.chkexe_] Error 1
make[4]: Leaving directory `/local/pbisbal/gcc-5.4.0/hdf5-1.10.0-patch1/test'
make[3]: *** [build-check-s] Error 2
make[3]: Leaving directory `/local/pbisbal/gcc-5.4.0/hdf5-1.10.0-patch1/test'
make[2]: *** [test] Error 2
make[2]: Leaving directory `/local/pbisbal/gcc-5.4.0/hdf5-1.10.0-patch1/test'
make[1]: *** [check-am] Error 2
make[1]: Leaving directory `/local/pbisbal/gcc-5.4.0/hdf5-1.10.0-patch1/test'
make: *** [check-recursive] Error 1
On 07/25/2016 04:31 PM, Miller, Mark C. wrote:
Yeah, I meant ‘kill -9’ not ctrl-C. I would expect ctrl-C to just abort the current command (e.g. ‘make check’). By killing the running process (lt-btree) alone, make check would just interpret that as some kind of failure and then move on to the next test.
Yes, you need a “true” parallel filesystem for any parallel tests. NFS will NOT work.
Running on /local makes a lot of sense other than for parallel. Whenever I run on NFS, things take a lot longer due to all the contention for that shared resource.
Mark
From: Hdf-forum <hdf-forum-bounces@lists.hdfgroup.org<mailto:hdf-forum-bounces@lists.hdfgroup.org>> on behalf of Prentice Bisbal <pbisbal@pppl.gov<mailto:pbisbal@pppl.gov>>
Reply-To: HDF Users Discussion List <hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>>
Date: Monday, July 25, 2016 at 1:22 PM
To: "hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>" <hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>>
Subject: Re: [Hdf-forum] How long should 'make check' take?
Where (e.g. on which kind of filesystem) are you trying to run ‘make check’?
Can you kill -9 that one process and see if it proceeds to completion on remaining tests?
Since I'm working on a serial build, right now, I'm using scratch disk local to the node I'm doing my builds on, mounted as '/local' When I do a control-C, the whole 'make check' process appears to end, it doesn't go on to the next test.
I'm now doing a build with minimal options specified in the configure step to see if maybe combinations of options was causing the problem.
For parallel tests, you *should* be using a parallel fileystem. I frequently forget about that.
Do you mean a shared filesystem, like NFS, or a true parallel filesystem like Lustre or GPFS? After realizing my stupidity of trying to do the parallel make check on my local disk, I did move it to an NFS filesystem.
Prentice
On 07/25/2016 04:07 PM, Miller, Mark C. wrote:
30+mins *does* seem a bit long.
Typically, a ‘make check’ for me completes *all* tests in a total of about 10-15 mins at the most.
Where (e.g. on which kind of filesystem) are you trying to run ‘make check’?
For parallel tests, you *should* be using a parallel fileystem. I frequently forget about that.
If the parallel filesystem is *extremely* busy and/or not setup to deal with small I/Os and/or files (as the tests generate), I could see there being *some* additional delay in completion.
The btree test I think is testing HDF5’s internal btree data structures for group objects. So, if its doing any I/O I think its a bunch of tiny HDF5 metadata stuff.
Can you kill -9 that one process and see if it proceeds to completion on remaining tests?
Mark
From: Hdf-forum <hdf-forum-bounces@lists.hdfgroup.org<mailto:hdf-forum-bounces@lists.hdfgroup.org>> on behalf of Prentice Bisbal <pbisbal@pppl.gov<mailto:pbisbal@pppl.gov>>
Reply-To: HDF Users Discussion List <hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>>
Date: Monday, July 25, 2016 at 12:59 PM
To: "hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>" <hdf-forum@lists.hdfgroup.org<mailto:hdf-forum@lists.hdfgroup.org>>
Subject: [Hdf-forum] How long should 'make check' take?
How long should make check normally take to run for a serial build?
After 'make check' was hanging for me at testphdf5 when building HDF5
with parallel support, make check now seems to be hanging when testing
btree2. It's been at the same spot for 30 minutes or more, but this
time, I see the process (lt-btree2) using CPU and memory in top, just
not much of it. Is this normal? I would expect file I/O tests to only
take a few minutes, not 30+ minutes.
--
Prentice
_______________________________________________
Hdf-forum is for HDF software users discussion.
Hdf-forum@lists.hdfgroup.org<mailto:Hdf-forum@lists.hdfgroup.org>
http://lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org
Twitter: https://twitter.com/hdf5
_______________________________________________
Hdf-forum is for HDF software users discussion.
Hdf-forum@lists.hdfgroup.org<mailto:Hdf-forum@lists.hdfgroup.org>http://lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org
Twitter: https://twitter.com/hdf5
_______________________________________________
Hdf-forum is for HDF software users discussion.
Hdf-forum@lists.hdfgroup.org<mailto:Hdf-forum@lists.hdfgroup.org>http://lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org
Twitter: https://twitter.com/hdf5
