Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
Ruby D

README.md

Deprecated. Use remote_syslog2 instead of this repo

remote_syslog Ruby daemon & sender

remote_syslog has been rewritten in Go as remote_syslog2. As a standalone binary, remote_syslog2 has fewer dependencies. It also depends on less code between the daemon and the OS.

Use remote_syslog2 instead of this repo.

Introduction

Lightweight Ruby daemon to tail one or more log files and transmit UDP syslog messages to a remote syslog host (centralized log aggregation).

remote_syslog generates UDP packets itself instead of depending on a system syslog daemon, so its configuration doesn't affect system-wide logging - syslog is just the transport.

Uses:

  • collecting logs from servers & daemons which don't natively support syslog
  • when reconfiguring the system logger is less convenient than a purpose-built daemon (e.g., automated app deployments)
  • aggregating files not generated by daemons (e.g., package manager logs)

The library can also be used to generate one-off log messages from Ruby code.

Tested with the hosted log management service Papertrail and should work for transmitting to any syslog server.

Installation

Install the gem, which includes a binary called "remote_syslog":

$ [sudo] gem install remote_syslog

Optionally, create a log_files.yml with the log file paths to read and the host/port to log to (see examples/log_files.yml.example). These can also be specified as command-line arguments (below).

Usage

Usage: remote_syslog [OPTION]... <FILE>...

Options:
    -c, --configfile PATH            Path to config (/etc/log_files.yml)
    -d, --dest-host HOSTNAME         Destination syslog hostname or IP (logs.papertrailapp.com)
    -p, --dest-port PORT             Destination syslog port (514)
    -D, --no-detach                  Don't daemonize and detach from the terminal
    -f, --facility FACILITY          Facility (user)
        --hostname HOST              Local hostname to send from
    -P, --pid-dir DIRECTORY          DEPRECATED: Directory to write .pid file in
        --pid-file FILENAME          Location of the PID file (default /var/run/remote_syslog.pid)
        --parse-syslog               Parse file as syslog-formatted file
    -s, --severity SEVERITY          Severity (notice)
        --strip-color                Strip color codes
        --tls                        Connect via TCP with TLS
        --tcp                        Connect via TCP (no TLS)
        --new-file-check-interval INTERVAL
                                     Time between checks for new files

Advanced options:
        --[no-]eventmachine-tail     Enable or disable using eventmachine-tail
        --debug-log FILE             Log internal debug messages
        --debug-level LEVEL          Log internal debug messages at level

Common options:
    -h, --help                       Show this message
        --version                    Show version

Example:
    $ remote_syslog -c configs/logs.yml -p 12345 /var/log/mysqld.log

Example

Typical:

$ remote_syslog

Daemonize and collect messages from files listed in ./config/logs.yml as well as the file /var/log/mysqld.log. Send to port logs.papertrailapp.com:12345:

$ remote_syslog -c configs/logs.yml -p 12345 /var/log/mysqld.log

Stay attached to the terminal, look for and use /etc/log_files.yml if it exists, write PID to /tmp/remote_syslog.pid, and send with facility local0 to a.example.com:514:

$ remote_syslog -D -d a.example.com -f local0 --pid-file /tmp/remote_syslog.pid /var/log/mysqld.log

Windows

Windows is not currently supported, though in certain situations it may work.

Auto-starting at boot

The gem includes sample init files, also available here. You may be able to:

$ cp examples/remote_syslog.init.d /etc/init.d/remote_syslog
$ chmod 755 /etc/init.d/remote_syslog

And then ensure it's started at boot, either by using:

$ sudo update-rc.d remote_syslog defaults

or by creating a link manually:

$ sudo ln -s /etc/init.d/remote_syslog /etc/rc3.d/S30remote_syslog

remote_syslog will daemonize by default.

Init files: remote_syslog.init.d (init.d), OS X launchd, supervisor, Ubuntu upstart

Optional: rvm

Remember that when using a Ruby version manager such as rvm, your interactive shell and init files need the version manager environment loaded. The rvm init.d instructions show how to create a wrapper script for the init files to run. A typical example is:

rvm wrapper ruby-1.9.3-p392 bootup remote_syslog

.. where ruby-1.9.3-p392 is the desired Ruby from rvm list. rvm will output the path to the new wrapper script which it created. Edit the init file to run the new wrapper script instead of running remote_syslog directly.

Sending messages securely

If the receiving system supports sending syslog over TCP with TLS, you can pass the --tls option when running remote_syslog:

$ remote_syslog --tls -p 1234 /var/log/mysqld.log

Important: remote_syslog depends on I/O code provided by the Ruby VM, eventmachine library, and OS. There is at least one environment and failure case where remote_syslog will not reconnect when using the --tls option. Although we've never been able to reproduce this problem, enough Papertrail customers have run into it that we'd suggest looking at alternative solutions. One of those is forwarding data to rsyslog and then using its TLS capabilities to log to Papertrail. For more information on that and other alternatives, please contact support@papertrailapp.com

Configuration

By default, the gem looks for a configuration in /etc/log_files.yml.

The gem comes with a sample config. Optionally:

$ cp examples/log_files.yml.example /etc/log_files.yml

log_files.yml has filenames to log from (as an array) and hostname and port to log to (as a hash). Wildcards are supported using * and standard shell globbing. Filenames given on the command line are additive to those in the config file.

Only 1 destination server is supported; the command-line argument wins.

files:
 - /var/log/httpd/access_log
 - /var/log/httpd/error_log
 - /var/log/mysqld.log
 - /var/run/mysqld/mysqld-slow.log
destination:
  host: logs.papertrailapp.com
  port: 12345

remote_syslog sends the name of the file without a path ("mysqld.log") as the syslog tag (program name). RFCs 3164 and 5424 limit the tag to 32 characters. Longer filenames are truncated to 32 characters.

After changing the configuration file, restart remote_syslog using the init script or by manually killing and restarting the process. For example:

/etc/init.d/remote_syslog restart

Advanced Configuration (Optional)

Here's an advanced config which uses all options.

Override hostname

Provide --hostname somehostname or use the hostname configuration option:

hostname: somehostname

Verify server certificate

Provide the public key for the remote host when using TLS:

ssl_server_cert: syslog.crt

Use a client certificate

Provide a client certificate when connecting via TLS:

ssl_client_cert_chain: syslog_client.crt
ssl_client_private_key: syslog_client.key

Detecting new files

remote_syslog automatically detects and activates new log files that match its file specifiers. For example, *.log may be provided as a file specifier, and remote_syslog will detect a some.log file created after it was started. Globs are re-checked every 10 seconds. Ruby's Dir.glob is used.

Note: messages may be written to files in the 0-10 seconds between when the file is created and when the periodic glob check detects it. This data is not currently acted on, though the default behavior may change in the future.

Also, explicitly-provided filenames need not exist when remote_syslog is started. remote_syslog can be pre-configured to monitor log files which are created later (or may never be created).

If globs are specified on the command-line, enclose each one in single-quotes ('*.log') so the shell passes the raw glob string to remote_syslog (rather than the current set of matches). This is not necessary for globs defined in the config file.

Log rotation

External log rotation scripts often move or remove an existing log file and replace it with a new one (at a new inode). The Linux standard script logrotate supports a copytruncate config option. With that option, logrotate will copy files, operate on the copies, and truncate the original so that the inode remains the same.

This comes closest to ensuring that programs watching these files (including remote_syslog) will not be affected by, or need to be notified of, the rotation. The only tradeoff of copytruncate is slightly higher disk usage during rotation, so we recommend this option whether or not you use remote_syslog.

Excluding files from being sent

Provide one or more regular expressions to prevent certain files from being matched.

exclude_files:
  - \.\d$
  - .bz2
  - .gz

Multiple instances

Run multiple instances to support more than one message-specific file format or to specify unique syslog hostnames.

To do that, provide an alternate PID path as a command-line option to the additional instance(s). For example:

--pid-file /var/run/remote_syslog_2.pid

Parse fields from log messages

Rarely needed. Usually only used when remote_syslog is watching files generated by syslogd (rather than by apps), like /var/log/messages.

remote_syslog can parse the program and hostname from the log line. When one file contains logs from multiple programs (like with syslog), the log line may include text that is not part of the log message, like a timestamp, hostname, or program name. remote_syslog will extract those and use them in the corresponding syslog packet fields.

To do that, use the config file option parse_fields with the name of a format supported by remote_syslog, or your own regex. Included format names are syslog and rfc3339. For example:

parse_fields: syslog

The included syslog format uses the regex (\w+ \d+ \S+) (\S+) ([^:]+): (.*) to parse standard syslog lines like this:

Jul 18 08:25:08 hostname programname[1234]: The log message

The included rfc3339 format uses the regex (\S+) (\S+) ([^: ]+):? (.*) to parse syslog lines with high-precision RFC 3339 timestamps, like this:

2011-07-16T08:25:08.651413-07:00 hostname programname[1234]: The log message

To parse a format other than those, provide your own regex. It should include 4 backreferences to parse, in order: timestamp, system name, program name, message.

Match and return empty strings for any empty positions where the log line doesn't provide a value. For example, given the log message:

something-meaningless The log message

One could use a regex to ignore "something-meaningless" (and not to extract a program or hostname). To ignore that prefix and return 3 empty values then the log message, use parse_fields with this regex:

parse_fields: "something-meaningless ()()()(.*)"

Per-file regexes are not supported. Run multiple instances with different config files.

Excluding lines matching a pattern

There may be certain log messages that you do not want to be sent. These may repetitive log lines that are "noise" that you might not be able to filter out easily from the respective application. To filter these lines, use the exclude_patterns with an array or regexes:

exclude_patterns:
 - exclude this
 - \d+ things

Prepending a string to log messages

Use prepend to prepend a string to every log message before transmitting. The string is prepended to the log message body, as if it occurred at the start of every log file line. Include a trailing space if desired.

Examples:

prepend: important: 

or:

prepend: cafebabe-1024-4096-badd-1234abcd1234 

Choosing app name

remote_syslog uses the log file name (like "access_log") as the syslog program name, or what the syslog RFCs call the "tag." This is ideal unless remote_syslog watches many files that have the same name.

In that case, tell remote_syslog to set another program name by creating symbolic link to the generically-named file:

cd /path/to/logs
ln -s generic_name.log unique_name.log

Point remote_syslog at unique_name.log. It will use that as the program name.

Troubleshooting

Installation

gem not found

Install a Ruby distribution, which typically takes a minute.

g++ not found

Install gcc and g++ so this system can compile C/C++ source. Installation is typically sudo yum install gcc-c++ (RPM-based distros) or sudo apt-get install build-essential (.deb-based distros).

Getting Encryption not available... or TLS is not supported...

The exact error might appear as:

Encryption not available on this event-machine

or

TLS is not supported by eventmachine installed on this system. The openssl-devel/openssl-dev package must be installed before installing eventmachine.

Install the OpenSSL C++ package for your distribution, then reinstall the eventmachine. For example:

  • .deb distros like Ubuntu: sudo apt-get install libssl-dev
  • RPM distros like Fedora sudo yum install openssl-devel

Then:

gem install eventmachine -f

Getting no such file to load -- mkmf (LoadError)?

Try these:

  • Ubuntu: determine which Ruby version is active with ruby -v.

    For 1.8.7: sudo apt-get install build-essential ruby1.8 ruby1.8-dev rubygems. For 1.9.x, including 1.9.1 and 1.9.3: sudo apt-get install build-essential ruby1.9.1-dev. For 2.0: sudo apt-get install build-essential ruby2.0-dev

  • Fedora: sudo yum install ruby-devel.

  • Getting errors about missing header files, like ssl.cpp? Try:
    • CentOS: sudo yum install libstdc++-devel ruby-devel

Getting Package ruby1.8 is not available...

The exact error might appear as:

Package ruby1.8 is not available, but is referred to by another package.` and/or `Package rubygems is not available, but is referred to by another package.` on Ubuntu 14.04

Ubuntu 14.04 changed the name of the ruby 1.8.7 packages. Try this instead:

sudo apt-get install build-essential ruby-full ruby

Freezes at the compilation stage

This can happen when the system is low on memory. The installation process starts up the compiler, but it gets killed as soon it consumes too much memory. Tailing /var/log/syslog or /var/log/messages will confirm whether or not this is occurring.

The solution is to temporarily stop any memory intensive tasks, install remote_syslog, and then restart them.

Operations

Reconnect failures

remote_syslog depends on I/O code provided by the Ruby VM, eventmachine library, and OS. There is at least one environment and failure case where remote_syslog will not reconnect when using the --tls option. Although we've never been able to reproduce this problem (and known occurrences are correspondingly rare), the dependency and problem are worth noting.

remote_syslog not found?

It may not be in your path. Run find / -name remote_syslog to locate it, then run it with the full path (such as /var/lib/gems/1.8/bin/remote_syslog).

The system rebooted and remote_syslog didn't start

Install an init file.

Logs not appearing?

Two commands are particularly useful for observing remote_syslog behavior. First, its own debugging:

remote_syslog --debug-level DEBUG --debug-log remote_syslog.log

This will write internal operations to the file remote_syslog.log.

Second, strace or ktrace shows the interaction between remote_syslog and the OS. To run strace against an existing remote_syslog instance (process ID 12345):

 strace -fp 12345 -s 500

Feel free to ask questions or report bugs.

Reporting bugs

  1. See whether the issue has already been reported: https://github.com/papertrail/remote_syslog/issues/
  2. If you don't find one, create an issue with a repro case.

Contributing

Once you've made your great commits:

  1. Fork remote_syslog
  2. Create a topic branch - git checkout -b my_branch
  3. Commit the changes without changing the Rakefile or other files unrelated to your enhancement.
  4. Push to your branch - git push origin my_branch
  5. Create a Pull Request or an Issue with a link to your branch
  6. That's it!
Something went wrong with that request. Please try again.