|--debug||Turn on MogileFS debug output.|
|--trackers=<[preferred_ip/]ip:port>[,<[preferred_ip/]ip:port>]*||Specify one or more trackers for your MogileFS installation. Note that you can specify preferred IPs to override the default IPs with. So it would look something like <B>10.10.0.1/10.0.0.1:8081B>.|
|--domain=<domain>||Set the MogileFS domain to use.|
|--class=<class>||Set the class within the domain to use. Defaults to _default.|
|--conf=<file>||Specify a configuration file to load from.|
|--lib=<directory>||Specify a directory to use as a library path. Right now, this should be the directory where you expect to find the MogileFS.pm file, if its not actually installed.|
inject|i Insert a resource into MogileFS. See INJECT OPTIONS and INJECT ARGUMENTS for the rest of how to use the inject mode. extract|x Extract a resource from MogileFS. See EXTRACT OPTIONS and EXTRACT ARGUMENTS for how to use extract. delete|rm Delete a resource. See DELETE OPTIONS and DELETE ARGUMENTS. locate|lo key List the paths to the file identified by the given key. list|ls List all big files contained in MogileFS. No options, no arguments. listkey|lsk key List all files which match the key. Key is just a prefix, and this will list all keys which match the prefix. So if you specify key as ABC1 then youll get all keys which start with the characters ABC1
The following options are used to control the behavior of the injector.
--bigfile|-b If specified, use chunking to break the resource into manageable pieces. --chunksize=<size>[B|K|M|G] When instructed to break files into chunks, the injector will use the specified chunk size as the maximum chunk size. Defaults to 64M. You can specify the chunk size manually and specify the unitsdefaults to bytes. --gzip|-z If specified, mogtool will gzip the data as its going into MogileFS. The resource will be marked as compressed.
Note that you do not need to specify this if the resource is already gzipped, but it doesnt hurt. (We automatically detect that and mark it as compressed.)
--overwrite If you previously were working on injecting a big file as chunks and the process died, normally mogtool refuses to do it again. Specify this option to force the overwrite of that file. --verify If on, we do a full MD5 verification of every chunk after it is replicated. This can take a while on large files! --description=<text> Specifies a description for this file. Optional, but assists in reporting and listing the large files in MogileFS. (This is also displayed in any receipts that are created.) --receipt=<email address>[, <email address>]* If specified, emails a copy of the receipt file to the specified comma-separated email addresses. Also creates a local filesystem copy of the receipt file. --concurrent=<number> Specifies the number of concurrent processes to run for MogileFS insertion. If you are noticing mogtool spend most of its time waiting for children and not actually buffering data, you may wish to raise this number. The default is 1 but weve found 3 or 4 work well.
resource What you actually want to inject. This can be a file, directory, or a raw partition in the format /dev/X.
Please see USAGE EXAMPLES for more information on how to inject these different types of resources and the differences thereof.
key Specifies the key to save this file to. For big files, the key is actually _big_N:key and key,# where N is one of a bunch of things we use and # is the chunk number.
Generally, you want this to be descriptive so you remember what it is later and can identify the file just by looking at the key.
--bigfile|-b If specified, indicates that this resource was chunked on injection and should be reassembled for extraction. --gzip|-z Specifies to mogtool that it should ungzip the output if and only if it was compressed when inserted into the MogileFS system. So, if youre extracting a file that wasnt gzipped to begin with, this doesnt do anything. --asfile Useful when extracting something previously inserted as a directorythis option instructs mogtool to treat the resource as a file and not actually run it through tar for decompression.
key Specifies the key to get the file from. destination What destination means varies depending on what type of resource youre extracting. However, no matter what, you can specify a single dash (<B>-B>) to mean STDOUT.
Please see the usage examples for more information on how extract works.
--bigfile|-b The resource is a big file and all chunks should be deleted.
key Specifies the key of the file to delete.
--verify Verify that the returned paths actually contain the file. The locate commands defaults to verify, you can disable it with --noverify. --bigfile|-b The resource is a big file and the locations of the information key should be printed.
key Specifies the key of the file to locate
0 Success during operation. 1 During the locate, list, or listkey operation, the key was not found. 2 Some fatal error occurred.
Please note that all examples assume you have a default config file that contains the tracker and domain to use. Saves us from having to clutter up the command line.
When it comes to using small files, mogtool is very, very easy.
$ mogtool inject foo.dbm foo.dbm.2004.12
Injects the file foo.dbm into MogileFS under the key of foo.dbm.2004.12.
$ mogtool inject --gzip foo.dbm foo.dbm.2004.12
Injects the same file to the same key, but compresses it on the fly for you.
$ mogtool extract foo.dbm.2004.12 newfoo.dbm
Retrieves the key foo.dbm.2004.12 and saves it as newfoo.dbm.
$ mogtool extract --gzip foo.dbm.2004.12 newfoo.dbm
$ mogtool extract foo.dbm.2004.12 -
Print the resource to standard out. If you want, you can pipe it somewhere or redirect to a file (but why not just specify the filename?).
Given mogtools ability to break files into chunks and later reassemble them, inserting large files (even files over the 4GB barrier) is relatively easy.
$ mogtool inject --bigfile largefile.dat largefile.dat
As expected, inserts the file largefile.dat into the MogileFS system under the name largefile.dat. Not very creative. Uses the default 64MB chunks.
$ mogtool inject --bigfile --chunksize=16M largefile.dat largefile.dat
Specify to use 16MB chunks instead of the default. Otherwise, the same.
$ mogtool inject --bigfile --chunksize=1000K --gzip largefile.dat somekey
Do it again, but specify 1000KB chunks, gzip automatically, and upload it under a different key somekey.
$ mogtool inject --bigfile --concurrent=5 --gzip largefile.dat somekey
Same as above, but use 5 children processes for uploading chunks to MogileFS. This can take up to 300MB of memory in this example! (It tends to use about (concurrency + 1) * chunksize bytes.)
$ mogtool inject --bigfile --chunksize=32M --concurrent=3 --gzip \ --receipt="firstname.lastname@example.org" --verify --description="A large file" \ largefile.dat somekey
Break this file into 128MB chunks, set a description, use 3 children to upload them, gzip the file as you go, do a full MD5 verification of every chunk, then email a receipt with all of the MogileFS paths to me.
Lots of flexibility with mogtool.
$ mogtool extract --bigfile somekey newfile.dat
In its basic form, extracts the previously inserted large file and saves it as newfile.dat.
$ mogtool extract --bigfile --gzip somekey newfile.dat
If the file was gzipped on entry, ungzip it and save the result. If it wasnt gzipped, then we just save it.
Directories are easily injected and extracted with mogtool. To create the data stream that is inserted into MogileFS, we use tar.
$ mogtool inject --bigfile mydir mykey
Run mydir through tar and then save it as mykey.
$ mogtool inject --bigfile --gzip --concurrent=5 mydir mykey
Inject, but also gzip and use multiple injectors.
Note how this is just like injecting a large file. See injection examples for large files for more examples.
$ mogtool extract --bigfile mykey .
Extract the previously injected directory mykey to your local directory.
$ mogtool extract --bigfile --asfile mykey foo.tar
Take the previously generated tarball and save it as foo.tar. Simply creates the file instead of extracting everything inside.
mogtool has the ability to inject raw partitions into MogileFS and to retrieve them later and write them back to a partition. Theyre treated just like directories for the most part, we just dont pipe things through tar.
$ mogtool inject --bigfile /dev/hda3 hda3.backup
Save a raw copy of your partition /dev/hda3 to the key hda3.backup.
$ mogtool inject --bigfile --gzip /dev/hda3 hda3.backup
Same, but compress on the fly during injection.
$ mogtool extract --bigfile hda3.backup /dev/hda4
<B>WARNING:B> Please make sure youre specifying the right parameter, as delete does not prompt for confirmation of the request!
$ mogtool delete thekey
Delete a normal file.
$ mogtool delete --bigfile thekey
Delete a chunked filethis deletes all chunks and the receipt, so the file is gone.
$ mogtool list backup
Lists all large files stored in MogileFS. It is not possible to list all normal files at this time.
$ mogtool listkey abc1
Lists all files in MogileFS whose keys start with the characters abc1.
Instead of adding a ton of options to the command line every time, mogtool enables you to create a default configuration file that it will read all of the options from. It searches two locations for a default configuration file: <B>~/.mogtoolB> and <B>/etc/mogilefs/mogtool.confB>. (Alternately, you can specify <B>--conf=whateverB> as an option on the command line.)
The file can consist of any number of the following items:
trackers = 10.0.0.3:7001, 10.10.0.5/10.0.0.5:7001 domain = mogiledomain class = fileclass lib = /home/foo/lib gzip = 1 big = 1 overwrite = 1 chunksize = 32M receipt = email@example.com, firstname.lastname@example.org verify = 1 concurrent = 3
None? Send me any you find! :)
--concurrent for extract It would be nice to have concurrent extraction going on. recover mode If the receipt file is ever corrupt in MogileFS it would be useful to recover a file given just a receipt. It would have the same arguments as the extract mode, except use a receipt file as the data source. partition size verification We can easily get the partition size when we save one to MogileFS, so we should use that information to determine during extraction if a target partition is going to be big enough. on the fly gzip extraction Right now we can gzip on an injection, but we should support doing decompression on the fly coming out of MogileFS. make list take a prefix If you can specify a prefix, that makes things easier for finding small files that are stored in MogileFS. more information on list Have list load up the info file and parse it for information about each of the big files being stored. Maybe have this as an option (-l). (This means the reading and parsing of info files should be abstracted into a function.)
Mark Smith <email@example.com> - most of the implementation and maintenance.
Brad Fitzpatrick <firstname.lastname@example.org> - concepts and rough draft.
Robin H. Johnson <email@example.com> - locate function.
Copyright (c) 2002-2004 Danga Interactive. All rights reserved.
Hey! <B>The above document had some coding errors, which are explained below:B>
Around line 289: Expected text after =item, not a number Around line 293: Expected text after =item, not a number
|perl v5.20.3||MOGTOOL (1)||2011-01-05|