Version 0.70, 10 July 2018
Author: Jeremy Elson [email protected]
Web page:
emlog is a Linux kernel module that makes it easy to access the most recent (and only the most recent) output from a process. It works just like "tail -f" on a log file, except that the storage required never grows. This can be useful in embedded systems where there isn't enough memory or disk space for keeping complete log files, but the most recent debugging messages are sometimes needed (e.g., after an error is observed).
The emlog kernel module implements simple character device driver. The driver acts like a named pipe that has a finite, circular buffer. The size of the buffer is easily configurable. As more data is written into the buffer, the oldest data is discarded. A process that reads from an emlog device will first read the existing buffer, then see new text as it's written, similar to monitoring a log file using "tail -f". (Non-blocking reads are also supported, if a process needs to get the current contents of the log without blocking to wait for new data.)
The current version of emlog should work under just about any Linux kernel in the 2.6.x (at least 2.6.32 and newer), 3.x, and 4.x series (at least up to 4.18-rc4).
emlog is free software, distributed under the GNU General Public License (GPL) version 2; see the file COPYING (in the distribution) for details.
If you want to compile emlog for use with the currently running kernel, simply run
makeOtherwise, you have to set either KVER (for linux kernel sources,
located in /lib/modules/<KVER>/build) or KDIR (for any other path):
make KDIR=/usr/src/linuxTwo files should be generated: the kernel module itself (emlog.ko),
and the nbcat utility that will be described later.
You can use them directly from the current directory or you can install them via
make installIf you chose to use emlog directly from the current directly, insert
the module into the kernel using the insmod command
insmod emlog.koOtherwise, modprobe should work:
modprobe emlogTo specify a different maximum buffer size limit:
modprobe emlog emlog_max_size=2048If successful, a message similar to
emlog:emlog_init: version 0.70 running, major is 251, MINOR is 1, max size 1024 K.
should show up in your kernel log (type dmesg to see it).
You can also verify that the module has been inserted by
typing lsmod or cat /proc/modules.
By default, a device file /dev/emlog is created (if you have devtmpfs
mounted onto /dev and/or have udev running) with a minimal allocated buffer.
It's ready to be written to/read from.
If you need more devices/buffers, you should can use the mkemlog program
to create device files that your processes can write to.
Usage mkemlog <logdevname> [size_in_kilobytes] [mode]
Create a log file with a 8k buffer with file permissions 0660
mkemlog /tmp/testlogCreate a log file with a 17k buffer with file permissions 0660
mkemlog /tmp/testlog 17Create a log file with a 12k buffer with file permissions 0644
mkemlog /tmp/testlog_12k 12 0644Create a log file with a 18k buffer with file permissions 0644, owned by a user with UID==1000.
mkemlog /tmp/testlog_18k 18 0644 1000The mkemlog requires the /dev/emlog file to be created.
If you do not have devtmpfs mounted onto /dev and/or have udev running,
Then you can manually create emlogs using mknod to create device
files that your processes can write to.
You need to know two numbers: the major and the minor.
You can find the major number by either of the following methods:
ls -l /dev/emlog
grep emlog /proc/devices
(source /sys/class/emlog/emlog/uevent ; echo "$MAJOR")
dmesg | grep emlogThe minor number is used to indicate the size of the ring buffer for that device file, specified as the the number of kilobytes (e.g., 1024 bytes). For example, to create an 8K buffer called 'testlog':
mknod /tmp/testlog c 251 8You can create as many devices as you like. Internally, emlog uses the file's inode and device numbers to identify the buffer to which the file refers. Note, that internal buffer size is currently limited to 128K.
Once the device file has been created, simply write to your device file as you would any normal named pipe, e.g.
echo hello > /tmp/testlogWrites to the log will never block because the buffer never runs out of space; old data is simply overwritten by new data.
You can read from the log in the normal way, e.g. using cat. By
default, reads block, just like tail -f, waiting for new log
data. For example:
cat /tmp/testlog
hello [we immediately see the hello that we wrote in the previous step]
_ [... and here's the cursor. the 'cat' process is now
blocked, waiting for new input. New data will be displayed
as it is written to the device by other processes.]
^C [use control-c, for example, to stop reading.]As of version 0.40, emlog's buffers can be read and/or monitored by multiple concurrent readers correctly. Data written to an emlog device will not disappear until it is overwritten by newer data, or the emlog module is removed. (In versions 0.30 and earlier, data was removed from the buffer the first time it was read.)
Type rmmod emlog or modprobe -r emlog to remove the emlog kernel
module and free all associated buffers. This won't work until all emlog
device files are closed.
-
emlog will allocate a fixed-size buffer on behalf of a device file if one of the following two conditions is true:
- A process has the file open for reading or writing
- A process has written text to the pipe
In other words, buffers are persistent, even after a process closes the emlog device. Therefore, it is possible (naturally) to fill virtual memory by creating many large emlog devices and writing one byte to all of them. Don't do that. All buffers will be freed when the emlog kernel module is removed.
-
Non-blocking reads work; i.e., setting O_NONBLOCK using ioctl() will cause an EAGAIN to be returned if there is no data ready. In addition, the select() and poll() functions will work correctly on emlog devices.
-
A small utility,
nbcat, is included with the emlog distribution. nbcat is similar tocat, but uses nonblocking reads. This utility can be used to copy the current contents of an emlog device without blocking to wait for more input. For example:nbcat /var/log/emlog-device-instance > /tmp/saved-log-file
...will copy the current contents of the named emlog device to a file
in /tmp.
Alternatively, you can use dd for that
dd if=/var/log/emlog-device-instance of=/tmp/saved-log-file bs=4096 iflag=nonblock 2>/dev/nullBy default, emlog creates only one device in /dev/emlog (or whereever
your devtmpfs is mounted) with minimal buffer size.
It doesn't make much sense to precreate devices with all possible buffer sizes.
emlog lets you create as many log devices as you like,
anywhere on the filesystem -- the module tells
them apart based on their inode number. Having a single log device
always exist in a single place (/dev) is much less useful.
Q: I'm seeing "I/O error" at a time other then when the module is inserted.
A: Oops - you've found a bug in emlog. Please report it.
Q: When I try to access an emlog device file for reading or writing, I get the error "no such device".
A: This probably means either that the emlog kernel module is not loaded; or, that the major number of the device file does not match the major number that emlog registered. To see which major number is being used by emlog, use any of the following methods:
grep emlog /proc/devices
(source /sys/class/emlog/emlog/uevent ; echo "$MAJOR")
dmesg | grep emlogQ: When I try to access an emlog device file for reading or writing, I get the error "invalid argument".
A: The minor number of the emlog device file must be a number
between 1 and 128, representing the number of kilobytes (1,024 bytes)
that should be used for emlog's ring buffer. Make sure you're
specifying a valid minor number in your mknod statement. Don't use
0.
Q: I see "no memory" errors when I try opening new emlog files.
A: Looks like you're out of virtual memory, sport.
Q: When I try to remove the emlog driver (rmmod emlog), I get the
error "Device or resource busy" or "rmmod: ERROR: Module emlog is in use".
A: That means a process is currently using an emlog device. You have
to wait until all processes close all emlog device files until the
driver can be removed. Try using lsof to see which files are in use
by which processes.
Q: I am trying to save a copy of the current emlog buffer to another
file, by typing cp /tmp/emlog-test /tmp/saved-log-copy, but cp just
sits there forever.
A: cp is blocked waiting for more data, just like cat does when
used with an emlog device. Use nbcat, the non-blocking cat utility
included with the emlog distribution; for example:
nbcat /tmp/emlog-test > /tmp/saved-log-copyQ: You've made my computer crash.
A: Sorry. If you can reproduce the problem I'll try to fix it.
- Racy einfo allocation/destruction.
This may cause memory leaks or crashes during concurrent opening of new emlog buffers,
or concurrent closing/opening of an emlog device.
(when loaded with
emlog_autofree=1(defaults to off))
Bug reports, patches, complaints, praise, and submissions of Central Services Form 27B/6, are welcomed at Emlog github page.
- Change the default size of /dev/emlog from 1KB to 256KB.
- Allow emlog devices to be up to 1MB large.
- Fixes for recent kernel / glibc, fix mkemlog.
- Allow to specify ownership of the log device.
- Add support for per e-info rwlock and add more debug. (fix nasty reader vs writer race condition)
- Allow dynamic sizing of emlog_max_size via module_param.
- Added mkemlog utility.
- Autofree module option (free associated buffer on last close).
- Create usable /dev/emlog by default (with non-zero sized buffer).
- Support for kernels >= 3.19.
- Dropped support for kernels < 2.6.20.
- Use pr_err() and friends instead of plain printk().
- Separate Kbuild file and makefile updates.
- Cleanup: types, static, etc.
- README conversion to Markdown syntax.
- Switched to char device region instead of a misc device.
- Support for both 2.6.x and 3.x kernels.
- Proper log-levels for printk().
- Reindented source code (converted tabs to spaces).
- stability fixes
- module init and remove rewritten
- Support for 3.0 kernel.
- Changed to misc device for auto-creation of /dev/emlog by udev
- fix build on 3.0.3 kernel
- auto register /dev/emlog by udev
- Updated to compile and work with 2.6.x kernels.
- replaced MODULE_PARM macro with module_param function
- 2.6 kernel functions update from Darien version.
- 2.6 Kernel Makefile
- Modified the emlog code to make it compatible with Linux 2.6 kernels.
- Concurrent readers and writers are now supported correctly (data is not consumed when it is first read, as it was in previous versions).
- emlog's ring buffers now allocated using vmalloc instead of kmalloc to avoid locking large blocks of contiguous physical memory.
- Added MODVERSIONS support
- Added the 'nbcat' utility - similar to cat, but does not block at the end of the data.
- Bug fix: both device number and inode number are now stored internally (instead of only the inode number). This prevents the (unlikely) possibility that emlogs on different filesystems might share a single buffer.
- Now compiles correctly for 2.4 series kernels.
- select() and poll() now work correctly on emlog devices.
- Bug fix: all instances should not share one wait queue!
- Initial public release.
Emlog was written by Jeremy Elson [email protected] at the University of Southern California's Information Sciences Institute as part of the SCADDS project http://www.isi.edu/scadds. SCADDS is an embedded systems research project. We use small PC/104-bus-based single-board-PCs using Linux. We wanted to save the debugging output from certain processes, but since these things have 16MB of disk space and 32MB of RAM, keeping complete log files was not an option. These tiny nodes do have serial ports running PPP, though, so it's possible to walk over to a node with a laptop, plug in a serial cable, and then telnet into the box. Using emlog, we can always keep the most recent debug messages from our processes; in case of an error, we can plug in a debug console and see what went wrong.
This work was supported by DARPA under grant No. DABT63-99-1-0011 as part of the SCADDS project, and was also made possible in part due to support from Cisco Systems.