search for: wdelay

Previously the file plugin supported ‘rdelay’ and ‘wdelay’ parameters for injecting delays (for testing) into read and write requests. This moves the functionality to a new delay filter so that it can be used with any plugin. --- TODO | 3 - configure.ac | 1 + filters/Makefile.am...

On 01/17/2018 04:38 PM, Eric Blake wrote: > On 01/17/2018 02:53 PM, Richard W.M. Jones wrote: >> Previously the file plugin supported ‘rdelay’ and ‘wdelay’ parameters >> for injecting delays (for testing) into read and write requests. This >> moves the functionality to a new delay filter so that it can be used >> with any plugin. >> --- > >> +/* Write data. */ >> +static int >> +delay_pwrite (struct nbdk...

On 01/17/2018 02:53 PM, Richard W.M. Jones wrote: > Previously the file plugin supported ‘rdelay’ and ‘wdelay’ parameters > for injecting delays (for testing) into read and write requests. This > moves the functionality to a new delay filter so that it can be used > with any plugin. > --- > +/* Write data. */ > +static int > +delay_pwrite (struct nbdkit_next *next, void *nxdata, >...

On Wed, Jan 17, 2018 at 08:01:35PM -0600, Eric Blake wrote: > Actually, thinking about this more: > > When I added zero support, I documented in commit ac3f294a that for the > file plugin, wdelay is indeed doubled on systems lacking efficient zero > support. But the fallback for handling EOPNOTSUPP is currently done at > the plugins.c level (ie. it is part of next->zero()) - so a filter > should never see a plugin returning EOPNOTSUPP. Or put another way, > your movement of...

...th reads and writes (for a good chance @@ -55,8 +56,8 @@ qemu-io -f raw -c "aio_write -P 1 0 512" -c "aio_write -P 2 512 512" \ # With --threads=1, the write should complete first because it was issued first nbdkit -v -t 1 -U - --filter=delay file test-parallel-file.data \ - wdelay=2 rdelay=1 --run 'qemu-io -f raw -c "aio_write -P 2 512 512" \ - -c "aio_read -P 1 0 512" -c aio_flush $nbd' | + wdelay=2 rdelay=1 --run 'timeout 10s </dev/null qemu-io -f raw \ + -c "aio_write -P 2 512 512" -c "aio_rea...

I have been running a local repository on a CentOS4 server. It serves CentOS4 clients (and FC5, FC6, and SuSE10.x if I bother to mention). My current options for exports is (ro,wdelay,insecure,root_squash). The ro and insecure options were added manually and others are default. I understand 'sync' is the default although it is not shown by exportfs -v. I also found this description in the NFS FAQ: "For the Linux implementation of NFS Version 3, using the "as...

On Wed, Jan 17, 2018 at 04:38:04PM -0600, Eric Blake wrote: > > +/* Zero data. */ > > +static int > > +delay_zero (struct nbdkit_next *next, void *nxdata, > > + void *handle, uint32_t count, uint64_t offset, int may_trim) > > +{ > > + write_delay (); > > + return next->zero (nxdata, count, offset, may_trim); > > If next->zero()

...e5 and http://pastebin.com/f6ff5890a I tried the same script on an USB-HDD and it works fine, even for / Here are some information about the NFS config: cat /proc/mounts | grep nfs: http://pastebin.com/f250c9f63 (I aslo tried bigger rsize and wsize) exportfs -v: /volume1 192.168.0.13(rw,wdelay,no_root_squash,no_subtree_check) cat /var/lib/nfs/etab: /volume1 192.168.0.13(rw,sync,wdelay,hide,nocrossmnt,secure,no_root_squash,no_all_squash,no_subtree_check,secure_locks,acl,anonuid=65534,anongid=65534) cat /var/lib/nfs/xtab: /volume1 molungus(rw,sync,wdelay,hide,nocrossmnt,s...

...may have spurious failures under heavy loads on the test machine, where # tuning the delays may help. -trap 'rm -f test-parallel-file.out' 0 1 2 3 15 - # With --threads=1, the write should complete first because it was issued first -nbdkit -v -t 1 -U - --filter=delay file file=file-data wdelay=2 rdelay=1 --run ' - $QEMU_IO -f raw -c "aio_write -P 2 1 1" -c "aio_read -P 1 0 1" -c aio_flush $nbd -' | tee test-parallel-file.out -if test "$(grep '1/1' test-parallel-file.out)" != \ -"wrote 1/1 bytes at offset 1 -read 1/1 bytes at offset 0&qu...

I tried reproducing the testsuite failure on test-parallel-*.sh using the same machine Rich posted a log from, but did not quickly hit it on a loop of make -j20 check TESTS=test-parallel-{file,nbd}.sh But I still think this series can't hurt. Eric Blake (2): nbd: Don't advertise writes if nbdkit is readonly tests: Make parallel tests more robust plugins/nbd/nbd.c | 2 ++

...ad in place on FC3 that worked. On the server, I have the following data structure: /data /data/archive /data/pictures /data/music Each is a separate ext3 filesystem mounted with rw,acl. /etc/exports presents all of these filesystems properly: # exportfs -v /data/pictures 172.17.17.0/27 (rw,wdelay,insecure,root_squash,no_subtree_check,anonuid=65534,anongid=65534) /data/archive 172.17.17.0/27 (rw,wdelay,insecure,root_squash,no_subtree_check,anonuid=65534,anongid=65534) /data/music 172.17.17.0/27 (rw,wdelay,insecure,root_squash,no_subtree_check,anonuid=65534,anongid=65534) /data...

...elp with following problem on centos 5.8 kernel and nfs4. On server i have bind mounted directory shared to the client with following setup: mounted directory: /srv/nfs/imagetest on /exports/imagetest type none (rw,bind) /etc/exports file: /exports 10.0.4.0/24(ro,insecure,sync,wdelay,no_subtree_check,no_root_squash,fsid=0,crossmnt) /exports/imagetest 10.0.4.20(rw,insecure,sync,wdelay,no_subtree_check,nohide,no_root_squash) On client I have mounted it with following command: 10.0.4.2:/ on /media/nfs type nfs4 (rw,nodev,sync,proto=tcp,retry=10,rsize=32768,wsize=32768...

Either patch 1 or patch 2 in isolation is sufficient to fix the problem that Rich forwarded on from an archlinux tester (name so I can credit them?). But both patches should be applied, as well as backported to appropriate stable branches, to maximize cross-version interoperability of nbdkit vs. plugins. Patch 3 will let us detect future similar bugs much faster. I want patch 4 to ensure that

...gt; milliseconds. Delay trim/discard operations by C<SECS> seconds or C<NN> milliseconds. +=item B<delay-extents=>SECS + +=item B<delay-extents=>NNB<ms> + +Delay block status (extents) operations by C<SECS> seconds or C<NN> +milliseconds. + =item B<wdelay=>SECS =item B<wdelay=>NNB<ms> diff --git a/filters/delay/delay.c b/filters/delay/delay.c index a00292b..d11958b 100644 --- a/filters/delay/delay.c +++ b/filters/delay/delay.c @@ -47,6 +47,7 @@ static int delay_read_ms = 0; /* read delay (milliseconds) */ static int delay_write_...

...d with writes longer than reads @@ -51,7 +57,7 @@ $QEMU_IO -f raw -c "aio_write -P 1 0 512" -c "aio_write -P 2 512 512" \ # With --threads=1, the write should complete first because it was issued first nbdkit -v -t 1 -U - --filter=delay file file=test-parallel-file.data \ - wdelay=2 rdelay=1 --run '$QEMU_IO -f raw -c "aio_write -P 2 512 512" \ + wdelay=2 rdelay=1 --run 'qemu-io -f raw -c "aio_write -P 2 512 512" \ -c "aio_read -P 1 0 512" -c aio_flush $nbd' | tee test-parallel-file.out if test &quot...

...d.au/19-Network_File_System.html#nfs4> and http://blog.laimbock.com/2009/05/21/nfsv4-on-centos-53-and-fedora-11/ and still can't see what i'm doing wrong {The Server} CentOS 5.4 64 Bit [root at inf1 /]# cat /etc/exports /nfs4exports *.scms.waikato.ac.nz(rw,insecure,sync,wdelay,no_subtree_check,no_root_squash,fsid=0,nohide) /nfs4exports/share *.scms.waikato.ac.nz(rw,insecure,sync,wdelay,no_subtree_check,no_root_squash,nohide) [root at inf1 /]# cat /etc/sysconfig/nfs # # Define which protocol versions mountd # will advertise. The values are "no" or "y...

This works OK on x86_64, but fails on our fast new Amberwing (aarch64) machine. I've attached the test-suite.log file, but I'm not very sure what's going wrong from that. Rich. -- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones Read my programming and virtualization blog: http://rwmj.wordpress.com virt-df lists disk usage of guests without needing to

This patch isn't complete (patch 6/6 isn't finished) so it's just for discussion, although it does compile and run. This introduces to nbdkit a concept of "filters" which can be placed in front of plugins to modify their behaviour. Some examples where you might use filters: * Serve a subset of the data, such as (offset, range) or a single partition from a disk image.

...t-delay-filter.pod index 3bfb039..62c0093 100644 --- a/filters/delay/nbdkit-delay-filter.pod +++ b/filters/delay/nbdkit-delay-filter.pod @@ -16,7 +16,7 @@ remote server, or to test certain kinds of race conditions in Linux. =head1 EXAMPLE - nbdkit --filter=delay file file=disk.img rdelay=100ms wdelay=100ms + nbdkit --filter=delay file disk.img rdelay=100ms wdelay=100ms =head1 PARAMETERS diff --git a/filters/error/nbdkit-error-filter.pod b/filters/error/nbdkit-error-filter.pod index 0f02282..76003c7 100644 --- a/filters/error/nbdkit-error-filter.pod +++ b/filters/error/nbdkit-error-filter.p...

...t-delay-filter.pod index 3bfb039..62c0093 100644 --- a/filters/delay/nbdkit-delay-filter.pod +++ b/filters/delay/nbdkit-delay-filter.pod @@ -16,7 +16,7 @@ remote server, or to test certain kinds of race conditions in Linux. =head1 EXAMPLE - nbdkit --filter=delay file file=disk.img rdelay=100ms wdelay=100ms + nbdkit --filter=delay file disk.img rdelay=100ms wdelay=100ms =head1 PARAMETERS diff --git a/filters/error/nbdkit-error-filter.pod b/filters/error/nbdkit-error-filter.pod index 0f02282..76003c7 100644 --- a/filters/error/nbdkit-error-filter.pod +++ b/filters/error/nbdkit-error-filter.p...