Activity for Mikhail T.
-
Mikhail T. posted a comment on ticket #40
The offset argument specifies where to begin in the file -- in case, you don't want to send all of it...
-
Patching the md5-code
-
The ChangeLog is stale -- not updated since 2012 -- the changes are, instead, enumerated in the ANNOUNCE. Could this be corrected, please?
-
Thank you for such a prompt reaction. It will take me a while to get to testing it though -- for now I'm still working with Dispy-4.10.5...
-
Your examples presume, the tasks are all homogeneous -- and a node either is or is not suitable for all of them. That's different from the problems I'm trying to solve, which both stem from the tasks being heterogeneous. Any node in my setup can process any task -- there is no need for any node to be excluded. But the tasks have different resource-requirements -- something a smart scheduler can help address by, for example, mixing heavy CPU-users together with heavy RAM-users. Currently, when determining,...
-
Feature-request: add different resource-consumption per task
-
Ok, then I've misunderstood the usage/purpose of the SharedJobCluster -- I only attempted to use it, when hitting the bug 19... Never mind then -- we don't even have the scheduler running. Sorry for the noise.
-
Ok, then I've misunderstood the usage/purpose of the SharedJobCluster -- I only attempted to use it, when hitting the bug 19... Never mind then -- we don't even have the scheduler running.
-
More specifically, this is our code using JobCluster. I wish, the same worked for SharedJobCluster, but it does not... def my_ip_addresses(family = socket.AF_INET): for i, nics in psutil.net_if_addrs().items(): for nic in nics: if nic.family == family: yield nic.address ... cluster = dispy.JobCluster( execTask, ... ip_addr = list(my_ip_addresses()) )
-
More specifically, this is our code using JobCluster. I wish, the same worked for SharedJobCluster, but it does not... def my_ip_addresses(family = socket.AF_INET): for i, nics in psutil.net_if_addrs().items(): for nic in nics: if nic.family == family: yield nic.address ... cluster = dispy.JobCluster( execTask, ... ip_addr = list(my_ip_addresses()) )
-
SharedJobCluster does not accept a list of IP-addresses in 4.10.x
-
If you don't mind, can you describe your use case? I forgot the details, but I recall, that, when I attempted to upgrade our Dispy PIP here 2+ years ago -- to 4.11.0 -- the existing code stopped working. I think, it had to do with the tasks remaining mutable in 4.10.x -- but in 4.11 they became private copies after cluster.submit() returns. Our code is written to retry any failed task once more. The failure-count is part of the task -- which became impossible in 4.11 -- that's my best recollection...
-
JobCluster.submit_node() broken in 4.10.6
-
Cannot write with 0.6.3
-
stat seems to have become much slower in 0.6.3
-
Tried 0.6.3 today. The lines are now counted correctly -- tail, actually, outputs 10 lines. But the -f flag is ineffective -- although the file continues to grow on the server, and clients accessing the server via NFS see the growth -- my FreeBSD client using smbnetfs does not detect the changes. Not until I restart tail.
-
Ubuntu may be following the same anti-developer principles of splitting packages, but the naming is different. The apt-get install librevenge-dev should work on a regular Ubuntu/Debian, not sure about the "Ubuntu on Windows"...
-
Unlike better operating systems (like FreeBSD), RedHat splits most third-party packages into foo and foo-devel. The former is sufficient to use the programs built by others, the latter is required to build new ones. Whenever I have use a RedHat/CentOS/Fedora machine, I hold my nose while adding the devel counterpart to each installed package. Only then does the system become usable for building from source -- which should be the only way to install open source software. So, Ian, what David was saying...
-
Tail is not working right...
-
Access to positional command-line arguments (argv)
-
Oh, I see. Well, maybe, introducing awk's concept of NF would help? As in: plot "foo.txt" using 1:(NF - 2)?
-
Counting fields from the end of the record
-
Camera/microphone settings should be verified before attempting a call
-
Thanks for the prompt response. The patches are too trivial for any credit to be necessary. You can attribute them to the "FreeBSD project" at large :) I can make my FreeBSD system available for test-builds. Let me know your /etc/passwd entry and ssh-key by e-mail.
-
Thanks for the prompt response. The patches are too trivial for any credit to be necessary. You can attribute them to the "FreeBSD project" at large :) I can make my FreeBSD system available for test-builds. Let me know your /etc/passwd entry and ssh-key via by e-mail.
-
Thanks for the prompt response. The patches are too trivial for any credit to be necessary. You can attribute them to the "FreeBSD project" at large :) I can make my FreeBSD system available for test-builds. Let me know your /etc/passwd entry and ssh-key via by e-mail.
-
Missing header-files
-
Porting to libfuse-3.x
-
Sorry, the 398e8169ea6d3d854af0173b1e66bf13c124f901 is not a fix -- there is no sys/xattr.h on FreeBSD either. If you think, the include is necessary at all -- have you tried compiling without it? -- then it should be protected by #ifdef __linux__, I think...
-
I used FreeBSD-11.1, but the iconv_open and friends are found in libc on FreeBSD-10.x as well -- not sure about 9.x, because I don't have any around.
-
There is no /opt/lib/samba4/lib/pkgconfig on my system, but invoking pkgconf --libs smbclient outputs the correct information without any special hacks. At any rate, the SMBCLIENT_LIBS should work to overwrite whatever pkgconf outputs, should not it? And it does not -- not quite...
-
Tried setting max_rw_block_size to 4194304 and then to 41943040 -- the performance dropped back down to mere hundreds of KB per second, direct_io or not.
-
By adding direct_io flag to the list of options of mount_fusefs, I was able to raise the throughput to 5-8 megabytes per second depending on the reader's options. Although a spectacular gain, this is still far below the 20MB/s reported by smbclient...
-
FWIW, I tried redefining the COMM_BUF_SIZE from 4096 to 16384, and even to 65536 -- but the throughput did not improve at all... Please, help.
-
smb_conn_no_reply -- empty, unused, but triggers a compiler warning
-
configure ignores SMBCLIENT_LIBS-variable
-
Very SLOW reads
-
FreeBSD has iconv-functions in libc
-
there is need to link Make that there is no need... Sorry...
-
No need to link with -lglib
-
clang complains about crash-triggering hack in list.h
-
attr/xattr.h is a Linux only "thing"
-
I was having the same problem and found this ticket with Google. The work-around I found is to put one of the options you'd like to pass to smbnetfs into the first column of the fstab-entry. For example, here is, what I have: -obig_writes /smb fusefs rw,failok,uid=101,late,mountprog=/opt/bin/smbnetfs (/opt is what I have for $LOCALBASE -- you, likely, have /usr/local). This works because the first field is passed by mount_fusefs to the specified program verbatim -- after the options. So the above...
-
Here is the test-log output...
-
Trying to upgrade the FreeBSD port to version 1.4.7 and the situation with tests seems even worse than before: Top-level Makefile invokes wish with argument all.tcl -- but such a script does not exist. The correct script-name seems to be all -- no suffix. Top-level Makefile forgets to cd into the tests/ subdirectory -- and, consequently, the test-scriptlets can't find their test-files A fair share of scriptlets still fails -- as reported originally. Not sure, if the failures are the same or not....
-
Trying to upgrade the FreeBSD port to version 1.4.7 and the situation with tests seems even worse than before: Top-level Makefile invokes wish with argument all.tcl -- but such a script does not exist. The correct script-name seems to be all -- no suffix Top-level Makefile forgets to cd into the tests/ subdirectory -- and, consequently, the test-scriptlets can't find their test-files * A fair share of scriptlets still fails -- as reported originally. Not sure, if the failures are the same or not....
-
I am having the same problem and found this ticket with Google. The work-around I found is to put one of the options you'd like to pass to smbnetfs into the first column of the fstab-entry. For example, here is, what I have: -obig_writes /smb fusefs rw,failok,uid=101,late,mountprog=/opt/bin/smbnetfs (/opt is what I have for $LOCALBASE -- you, likely, have /usr/local). This works because the first field is passed by mount_fusefs to the specified program verbatim -- after the options. So the above...
-
I am having the same problem and found this ticket with Google. The work-around I found is to put one of the options you'd like to pass to smbnetfs into the first column of the fstab-entry. For example, here is, what I have: -obig_writes /smb fusefs rw,failok,uid=101,late,mountprog=/opt/bin/smbnetfs (/opt is what I have for $LOCALBASE -- you, likely, have /usr/local). This works because the first field is passed by mount_fusefs to the specified program verbatim -- after the options. So the above...
-
FAAD2_VERSION remains "unknown"
-
Yep, I tried the same config from my home computer and connected promptly. It still does not work from my office machine -- the network admin confirmed, that GRE is blocked and will be making the necessary changes soon (hopefully).
-
Yep, I tried the same config from my home computer and connected promptly. The office machine does not work -- the network admin confirmed, that GRE is blocked and will be making the necessary changes soon (hopefully).
-
Thanks, Eugene! I'll be discussing this with our network admin. Meanwhile, do you think, the daemon could've provided some more explicit diagnostics? Like "haven't received GRE response from peer within %d"?
-
Could you share your config, please (without the auth, obviously)? Whatever I try, I remain stuck with the L1 showing device pptp "DOWN"... The log says: Nov 3 16:40:33 dufus mpd: process 25375 started, version 5.8 (mi@dufus 13:53 3-Nov-2017) Nov 3 16:40:33 dufus mpd: CONSOLE: listening on 127.0.0.1 5005 Nov 3 16:40:33 dufus mpd: web: listening on 0.0.0.0 5006 Nov 3 16:40:33 dufus mpd: [B1] Bundle: Interface ng0 created Nov 3 16:40:33 dufus mpd: [L1] Link: OPEN event Nov 3 16:40:33 dufus mpd: [L1]...
-
Could you share your config, please (without the auth, obviously)? Whatever I try, I remain stuck with the L1 showing device pptp "DOWN"... Thank you!
-
Could you share your config, please (without the auth, obviously)? Thank you!
-
Error-checking is lacking
-
Add const-poisoning
-
It is "external_...", rather than "ext_...", is not it?
-
Thanks for the explanation... I asked, because I encountered a similar crash in FreeImageTag.cpp,...
-
Thanks for the explanation... I asked, because I encountered a similar crash in FreeImageTag.cpp...
-
Should not try have caught the crash, however?
-
Міхаіл and/or Hervé! Could you post the actual patch against 3.17, that addresses...
-
Are bundled self-tests supposed to work?
-
Maybe, it is time for a new release? jxgrabkey-0.3.3.tar.xz?
-
<unistd.h> is required for usleep()
-
Actually, trying to build (and test) it now on CentOS-6 (with OpenJDK-1.8) I observe...
-
Actually, trying to build (and test) it now on CentOS-6 (with OpenJDK-1.8) I observe...
-
Actually, trying to build (and test) it now on CentOS-6 (with OpenJDK-1.8) I observe...
-
Well, you could add setting of LD_LIBRARY_PATH to your build.xml, but it does not...
-
Well, you could add setting of LD_LIBRARY_PATH to your build.xml, but it does not...
-
I found the patch necessary for the FreeBSD port of JXGrabKey to be able to perform...
-
Argh... Sorry about formatting: --- JXGrabKey/C++/src/JXGrabKey.cpp 2010-02-28 10:11:49.000000000...
-
XKeycodeToKeysym is deprecated
-
Use System.loadLibrary instead of System.load
-
pthread_spin_init used incorrectly
-
Test tlsIO-2.11 fails SOMETIMES
-
Fix the test/format_test/format_checks.sh.in
-
Fix compiler problems to allow builds with clang++
-
When I click on "Create Ticket" here, I see the banner. Shot of window attached....
-
No place to file Gallery 3 bugs
-
After recovering data, tsk_recover goes into some infinite loop
-
mtime, other attributes aren't preserved by tsk_recover
-
Tests seem to hang
-
TLS1.1 and TLS1.2 out to be off if not enabled at compile-time
-
Here is my collection of fixes -- adapted for 1.6.4. The unused enum-values need...
-
Here is my collection of fixes -- adapted for 1.6.4. The unused enum-values need...
-
Try this patch...
-
The patch, that allows the program to proceed regardless of signature-verification...
-
The patch, that allows the program to proceed regardless of signature-verification...
-
Not sure, if this is, what I need, but the text encoding for Tag 1 (ID3v1) is set...
-
Copying "tag 2" into "tag 1" broken
-
Based on the available information my best guess is a compiler/platform problem which...
-
I for one like -Werror -- because it helps identify the bugs like this one. The bugs...
-
I think, the difference is in using the array notation ([]) rather than pointer (*)....
-
Redundant "const" breaks build with a picky compiler
-
Thank you, Stefan, for the prompt response. can you login with Pidgin/SIPE successfully...
1 >