This release enables lessfs background truncation tasks to resume after an umount or an unexpected shutdown. Background truncation is now enabled by default. Since lessfs uses the freelist database to store the state of truncation this database now needs to be configured in lessfs.cfg even when the tc datastore is used.
-
Archives
- January 2015
- August 2014
- February 2014
- December 2013
- May 2013
- March 2013
- January 2013
- December 2012
- July 2012
- June 2012
- April 2012
- January 2012
- October 2011
- September 2011
- August 2011
- June 2011
- May 2011
- April 2011
- March 2011
- February 2011
- January 2011
- December 2010
- October 2010
- September 2010
- August 2010
- July 2010
- June 2010
- April 2010
- March 2010
- February 2010
- January 2010
- December 2009
- November 2009
- October 2009
- September 2009
- August 2009
- June 2009
- March 2009
-
Meta
Hi, I am sorry to trouble you, everybody
When I installed lessfs software in my computer, I use the following command:
1. mkdir /data
2. cd /data
3. mkdir dta mta
then:
4. du -sh /data
16K .
and:
5. ./mklessfs ete/lessfs.cfg
and then:
6. du -sh /data
73M .
then use the command :
7. ./lessfs ete/lessfs.cfg /fuse
finally I cp five files from ./mydata/ to /fuse/
8. cp a.vdi b.vdi c.vdi d.vdi e.vdi /fuse/
( notice: b.vdi, c.vdi, d.vdi and e.vdi are all the copys of a.vdi, and the size of a.vdi is 1.6G )
after about 3 minutes, the work is completed
and then :
9. du -sh /data
the result is : 989M
Now I want to know:
/data 中的数据：dta/ mta/ 是 最终去重得到的结果吗？
the data in /data/ directory (/data/dta/ /data/mta/) is the result of operation after dedup ? That is, after dedup operation, my five files (about 7G) is deduped in 989M ?
Thank you!
the following is the information in lessfs.stats:
[root@localhost .lessfs]# cat lessfs_stats
INODE SIZE COMPRESSED_SIZE FILENAME
10 0 0 lessfs_stats
14 0 0 enabled
15 0 0 backlog
16 1683005952 954985266 a.vdi
17 1683005952 0 b.vdi
18 1683005952 0 c.vdi
19 1683005952 0 d.vdi
20 1683005952 0 e.vdi
Yes. dta/ directory holds (deduplicated & compressed) blocks of your files, while information how to reassemble them back in your files is stored in mta/ directory.
Hi,
I think your software is really neat. I have about 800GB taking up only 140GB and another where 400GB is taking up around 200GB.
With the background truncation, does this prevent database corruption in tokyo cabinet if power is lost to the machine during a write?
Although I have been unable to corrupt TC with power outages with the latest lessfs releases, database damage can always occur. Replication or (snapshot) backups are the only way to safeguard you from these events.
First, Thank u for your help, but i still donnot know whether what I do is right.
There is no corruption because the power is always on!
By the way, Can you tell me which function or functions in lessfs do the operation of dedup, thank u very much.
Thank you for all the great job
I’ve posted on the ML but it seems almost dead… I’ve built a small Munin monitor for lessfs, that you can get here:
http://update.intellique.com/pub/munin-lessfs-0.1.tgz
Debian version :
http://update.intellique.com/pub/munin-lessfs_0.1_all.deb
On other aspects, I could send some patches to enhance lessfs (for instance providing error messages when mkfs fail
how should I proceed?
For now please feel free to email patches directly to me.
I’ll be happy to merge them with the next release.
Thanks in advance,
Mark
P.S. The mailing list that you are referring to is : lessfs@googlegroups.com?
No, I’ve used the sourceforgelessfs-user mailing list. I’ll use the google group then, thank you
BTW I’ve set up a 19 TB system that I’m currently filling up (that’s quite slow…) to stress lessfs capabalities. It hosts 4 TB used space ( 7 TB of actual data) at the moment, filling at about 500GB/day. I’ll keep you informed in case of any trouble (all fine so far).