Marc Fournier [Wed, 3 Dec 2014 15:15:28 +0000 (16:15 +0100)]
netlink: ensure size_t portability in DEBUG() statement
This prevents the following error when building on 32bit systems with
gcc 4.7.2:
netlink.c: In function 'ir_read':
netlink.c:783:7: error: format '%lu' expects argument of type 'long unsigned int', but argument 5 has type 'size_t' [-Werror=format]
When reading from tables, upon errors the PDUs sent are already
freed by snmp_synch_response since they are right after
snmp_send is called.
This commit syncs collectd's approach with other occurences of
snmp_synch_response calls.
There might be a few corner cases where we leak PDUs, but it
is unclear how to check for those since we would need to
have an indication that snmp_send was never called, which
as far as I can tell is not possible.
The potential for failure in snmp_send is rather low and will
be easily spotted though, since when crafting invalid PDUs
snmp send will constantly fail and since valid configurations
can never leak memory.
When reading from tables, upon errors the PDUs sent are already
freed by snmp_synch_response since they are right after
snmp_send is called.
This commit syncs collectd's approach with other occurences of
snmp_synch_response calls.
There might be a few corner cases where we leak PDUs, but it
is unclear how to check for those since we would need to
have an indication that snmp_send was never called, which
as far as I can tell is not possible.
The potential for failure in snmp_send is rather low and will
be easily spotted though, since when crafting invalid PDUs
snmp send will constantly fail and since valid configurations
can never leak memory.
Vincent Bernat [Fri, 7 Nov 2014 14:20:22 +0000 (15:20 +0100)]
network: don't enable gcrypt thread callbacks when gcrypt recent enough
From `gcrypt.h`:
> NOTE: Since Libgcrypt 1.6 the thread callbacks are not anymore used.
> However we keep it to allow for some source code compatibility if used
> in the standard way.
Otherwise, we get a deprecation warning which is turned into an error:
```
CC libcollectdclient_la-network_buffer.lo
../../../src/libcollectdclient/network_buffer.c:58:15: error: 'gcry_thread_cbs' is deprecated (declared at /usr/include/gcrypt.h:213) [-Werror=deprecated-declarations]
GCRY_THREAD_OPTION_PTHREAD_IMPL;
```
When running f3706b0b87, the following command gets sent to redis:
"ZADD" "collectd/hostname/entropy/entropy" "1415487432.000000" "1415487432:932"
Meaning the value actually stored, and later returned by redis is:
"<timstamp>:<value>".
b7984797 accidentally dropped the comma separating the timestamp and the
value, which leads the plugin to store a somewhat confusing value in
redis:
"ZADD" "collectd/hostname/entropy/entropy" "1415487432.000000" "1415487432932"
Syslog: if we can't find the loglevel specified by the configuration string default to 'info' and warn about the unknown configuration option. no way to make syslog totaly silent anymore.
curl_json: fix the array access implemented in f1e1e37e
In the avl-tree we store two different structs, cj_key_t for the value
we search, c_avl_tree_t for sub-nodes.
The old version does assume when it will find a key, and when a tree,
which doesn't have to be right in all cases.
Therefore we utilize the magic to revalidate this cast.
Being able to tell tree from key, we now can also implement array access
on the right most node of the tree-path.
Sebastian Harl [Sun, 19 Oct 2014 18:25:50 +0000 (20:25 +0200)]
Use the complain mechanism to report write failures.
This ensures that we don't spam the log for each submitted value. Instead, an
exponential backup is used to report permanent errors starting at one message
per second.
This code would throw an error for some systems, because the temperature
couldn't be read. Since the plugin still works in general, this was very
confusing to users. Becaus the temperature, even when read sucessfully, was
never dispatched, remove the code entirely. We might want to add this back to
the apple_sensors plugin eventually.
This moves the socket creation logic so it's called from
networt_send_buffer_plain(). This allows us to recover after network
failures or when collectd was started before the network was available.
This is called from the curl and memcachec plugins. Previously, gauge
average, minimum and maximum values were calculated since the start of
the daemon, rather than per-interval. This is not the documented
functionality.
Florian Forster [Thu, 28 Aug 2014 18:41:25 +0000 (20:41 +0200)]
java plugin: Make sure cjni_thread_detach() is called on all paths.
Each call to cjni_thread_attach() much be accompanied by
a call to cjni_thread_detach(). Some error handling cases were missing
the call, potentially locking the plugin up.
Florian Forster [Wed, 27 Aug 2014 11:04:59 +0000 (13:04 +0200)]
collectd.conf(5): Improve documentation of LoadPlugin.
Mention AutoLoadPlugin in the relevant places. Add a <Plugin /> block
for the "df" plugin to make it easier to spot that it is preceded by a
<LoadPlugin /> *block*.
Sebastian Harl [Sun, 27 Jul 2014 12:15:23 +0000 (14:15 +0200)]
utils_vl_lookup: Fixed a race when creating user objects.
This could cause multiple aggregation instances to be created in the
aggregation plugin when first writing data to the plugin. This, in turn, led
to "value too old" warnings because subsequently all data was submitted twice.
Thanks to @faxm0dem for reporting this in GH #535.
Plugin exec: change notification time to integer.
Acording to collectd-exec.5 NOTIFICATION DATA FORMAT: time is epoch, i. e. as seconds since 1970-01-01 00:00:00 UTC.
Configparser: when we alocate an empty list, we also need to reset the counter; else we will trip over this later.
This fixes https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=750440
projects / thirdparty / collectd.git / log
