Configured by setting mds_dir_max_commit_size in conf, or else
by looking at osd_max_write_size. This should lead to sane
max commits even if the user doesn't specify anything.
This will be used in the next commit or to by CDir.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Instead, commit if a certain percentage of the dentries are dirty.
Configurable via mds_dir_commit_ratio!
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
send_linger always creates a new Op, but op_submit does not fill in
the target pg if an existing session is passed in, so when a linger
was resent, it had the wrong pg set.
This caused a crash in cosd with debugging turned on when running
testlibrbd twice. This occurred because the object context for the
linger in the wrong pg had no object name set.
Currently, we haven't read the configuration at the time we initialize
these locks. So we can't know whether lockdep has been enabled, or what
verbosity it is supposed to have. So just disable it on these locks.
Potentially ExportControl's initialization could be moved to after
g_conf.lockdep and g_conf.debug_lockdep have been read from the
configuration, if lockdep is needed for this component.
ConfFile probably doesn't need a lock at all, but that's another story.
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
Configured by setting mds_dir_max_commit_size in conf, or else
by looking at osd_max_write_size. This should lead to sane
max commits even if the user doesn't specify anything.
This will be used in the next commit or to by CDir.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Instead, commit if a certain percentage of the dentries are dirty.
Configurable via mds_dir_commit_ratio!
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Fix
$ ./ceph -m 10.0.1.252:6790 health
2011-02-11 12:30:38.132728 mon <- [health]
2011-02-11 12:30:38.133243 mon <- [health]
2011-02-11 12:30:38.133697 mon0 -> 'Access denied' (-13)
when the IP we specify is, e.g., for mon.b.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Use safe_write. Check error code of close.
Fix what looks like a "lack of braces around multi-statement block"
error where we always complain that we're unable to write the pid file,
even if we succeed.
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
Create safe_read_exact, which fails with -EDOM unless it reads exactly
the requested number of bytes. More often, what we want is the
semantics of safe_read, which returns the number of bytes it read, or
negative on error. Likewise for pread.
Write doesn't need variants because failing to write all the bytes is
always an error.
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
