-<p>I spent Monday and tuesday this week in London with a lot of the
-people involved in the boot system on Debian and Ubuntu, to see if we
-could find more ways to speed up the boot system. This was an Ubuntu
-funded
-<a href="https://wiki.ubuntu.com/FoundationsTeam/BootPerformance/DebianUbuntuSprint">developer
-gathering</a>. It was quite productive. We also discussed the future
-of boot systems, and ways to handle the increasing number of boot
-issues introduced by the Linux kernel becoming more and more
-asynchronous and event base. The Ubuntu approach using udev and
-upstart might be a good way forward. Time will show.</p>
-
-<p>Anyway, there are a few ways at the moment to speed up the boot
-process in Debian. All of these should be applied to get a quick
-boot:</p>
-
-<ul>
-
-<li>Use dash as /bin/sh.</li>
-
-<li>Disable the init.d/hwclock*.sh scripts and make sure the hardware
- clock is in UTC.</li>
-
-<li>Install and activate the insserv package to enable
- <a href="http://wiki.debian.org/LSBInitScripts/DependencyBasedBoot">dependency
- based boot sequencing</a>, and enable concurrent booting.</li>
-
-</ul>
-
-These points are based on the Google summer of code work done by
-<a href="http://initscripts-ng.alioth.debian.org/soc2006-bootsystem/">Carlos
-Villegas</a>.
-
-<p>Support for makefile-style concurrency during boot was uploaded to
-unstable yesterday. When we tested it, we were able to cut 6 seconds
-from the boot sequence. It depend on very correct dependency
-declaration in all init.d scripts, so I expect us to find edge cases
-where the dependences in some scripts are slightly wrong when we start
-using this.</p>
-
-<p>On our IRC channel for this effort, #pkg-sysvinit, a new idea was
-introduced by Raphael Geissert today, one that could affect the
-startup speed as well. Instead of starting some scripts concurrently
-from rcS.d/ and another set of scripts from rc2.d/, it would be
-possible to run a of them in the same process. A quick way to test
-this would be to enable insserv and run 'mv /etc/rc2.d/S* /etc/rcS.d/;
-insserv'. Will need to test if that work. :)</p>
+<p>My file system sematics program
+<a href="http://people.skolelinux.org/pere/blog/Testing_if_a_file_system_can_be_used_for_home_directories___.html">presented
+a few days ago</a> is very useful to verify that a file system can
+work as a unix home directory,and today I had to extend it a bit. I'm
+looking into alternatives for home directory access here at the
+University of Oslo, and one of the options is sshfs. My friend
+Finn-Arne mentioned a while back that they had used sshfs with Debian
+Edu, but stopped because of problems. I asked today what the problems
+where, and he mentioned that sshfs failed to handle umask properly.
+Trying to detect the problem I wrote this addition to my fs testing
+script:</p>
+
+<pre>
+mode_t touch_get_mode(const char *name, mode_t mode) {
+ mode_t retval = 0;
+ int fd = open(name, O_RDWR|O_CREAT|O_LARGEFILE, mode);
+ if (-1 != fd) {
+ unlink(name);
+ struct stat statbuf;
+ if (-1 != fstat(fd, &statbuf)) {
+ retval = statbuf.st_mode & 0x1ff;
+ }
+ close(fd);
+ }
+ return retval;
+}
+
+/* Try to detect problem discovered using sshfs */
+int test_umask(void) {
+ printf("info: testing umask effect on file creation\n");
+
+ mode_t orig_umask = umask(000);
+ mode_t newmode;
+ if (0666 != (newmode = touch_get_mode("foobar", 0666))) {
+ printf(" error: Wrong file mode %o when creating using mode 666 and umask 000\n",
+ newmode);
+ }
+ umask(007);
+ if (0660 != (newmode = touch_get_mode("foobar", 0666))) {
+ printf(" error: Wrong file mode %o when creating using mode 666 and umask 007\n",
+ newmode);
+ }
+
+ umask (orig_umask);
+ return 0;
+}
+
+int main(int argc, char **argv) {
+ [...]
+ test_umask();
+ return 0;
+}
+</pre>
+
+<p>Sure enough. On NFS to a netapp, I get this result:</p>
+
+<pre>
+Testing POSIX/Unix sematics on file system
+info: testing symlink creation
+info: testing subdirectory creation
+info: testing fcntl locking
+ Read-locking 1 byte from 1073741824
+ Read-locking 510 byte from 1073741826
+ Unlocking 1 byte from 1073741824
+ Write-locking 1 byte from 1073741824
+ Write-locking 510 byte from 1073741826
+ Unlocking 2 byte from 1073741824
+info: testing umask effect on file creation
+</pre>
+
+<p>When mounting the same directory using sshfs, I get this
+result:</p>
+
+<pre>
+Testing POSIX/Unix sematics on file system
+info: testing symlink creation
+info: testing subdirectory creation
+info: testing fcntl locking
+ Read-locking 1 byte from 1073741824
+ Read-locking 510 byte from 1073741826
+ Unlocking 1 byte from 1073741824
+ Write-locking 1 byte from 1073741824
+ Write-locking 510 byte from 1073741826
+ Unlocking 2 byte from 1073741824
+info: testing umask effect on file creation
+ error: Wrong file mode 644 when creating using mode 666 and umask 000
+ error: Wrong file mode 640 when creating using mode 666 and umask 007
+</pre>
+
+<p>So, I can conclude that sshfs is better than smb to a Netapp or a
+Windows server, but not good enough to be used as a home
+directory.</p>
+
+<p>Update 2010-08-26: Reported the issue in
+<a href="http://bugs.debian.org/594498">BTS report #594498</a></p>
+
+<p>Update 2010-08-27: Michael Gebetsroither report that he found the
+script so useful that he created a GIT repository and stored it in
+<a href="http://github.com/gebi/fs-test">http://github.com/gebi/fs-test</a>.</p>
projects / homepage.git / blobdiff