Flushing file writes to disk with 100% reliability
in [Unix Programming]
|
Prev: Learn about proxy sites and how to use them to open blocked sites unlimited downloads from RapidShare and megaupload and increase the speed of the Internet with new sites for free
Next: How reliable are named pipes?
From: Jens Thoms Toerring on 8 Apr 2010 22:12 In comp.unix.programmer Peter Olcott <NoSpam(a)ocr4screen.com> wrote: > I was told to use fsync() not sync(). fsync(2) is the C function to use while sync(1) is a "program" you can invoke from the command line (or from a script) to do the same (well, fsync() is for a single file descriptor while sync() takes care of all disks...). Just have a look at the man pages for both of them - the fsync(2) Linux man page also mentions some caveats. (And the man page for open(2) has some bits of information about O_SYNC you have been asking about). Regards, Jens PS: The xxxx(y) notation means that the function/program 'xxxx' is in section 'y' of the man pages. Thus to get the descrip- tion for the 'xxxx(y)' for sure you may have to use man y xxxxx This is important for cases where functions/programs of the same name exist and allows to find what you're actually loo- king for. E.g. there is a 'write' utility - which you may get when you just do man write But when you look for the systems write() function you thus may have to use man 2 write and thus this write() is often referred to as write(2). -- \ Jens Thoms Toerring ___ jt(a)toerring.de \__________________________ http://toerring.de
From: Ian Collins on 8 Apr 2010 22:20 On 04/ 9/10 01:58 PM, Peter Olcott wrote: > > Does the MySQL InnoDB storage engine have a journal file > like SQLite for crash recovery? I suggest you do some background reading: http://dev.mysql.com/doc/refman/5.5/en/innodb.html To quote the first paragraph: "InnoDB is a transaction-safe (ACID compliant) storage engine for MySQL that has commit, rollback, and crash-recovery capabilities to protect user data." -- Ian Collins
From: Peter Olcott on 8 Apr 2010 22:37 "Ian Collins" <ian-news(a)hotmail.com> wrote in message news:827evjFp8jU20(a)mid.individual.net... > On 04/ 9/10 01:21 PM, Peter Olcott wrote: >> "Ian Collins"<ian-news(a)hotmail.com> wrote in message >> news:827d0gFp8jU18(a)mid.individual.net... >>> On 04/ 9/10 12:43 PM, Peter Olcott wrote: >>>> "Ian Collins"<ian-news(a)hotmail.com> wrote: >> Exactly how fault tolerant is it with the server's power >> cord yanked from the wall? > > As good as any. If you want 5 nines reliability you have > to go a lot further than synchronous writes. My main > server has highly redundant raid (thinks to ZFS), > redundant PSUs and a UPS. I'm not quite at the generator > stage yet, our power here is very dependable :). Do you also have disk write caching turned off?
From: Peter Olcott on 8 Apr 2010 22:38 "Ian Collins" <ian-news(a)hotmail.com> wrote in message news:827h6jFp8jU22(a)mid.individual.net... > On 04/ 9/10 01:58 PM, Peter Olcott wrote: >> >> Does the MySQL InnoDB storage engine have a journal file >> like SQLite for crash recovery? > > I suggest you do some background reading: > > http://dev.mysql.com/doc/refman/5.5/en/innodb.html > > To quote the first paragraph: > > "InnoDB is a transaction-safe (ACID compliant) storage > engine for MySQL that has commit, rollback, and > crash-recovery capabilities to protect user data." > > -- > Ian Collins OK that sounds good. Do you also have to turn off disk write caching?
From: Ian Collins on 8 Apr 2010 22:47
On 04/ 9/10 02:38 PM, Peter Olcott wrote: > "Ian Collins"<ian-news(a)hotmail.com> wrote in message > news:827h6jFp8jU22(a)mid.individual.net... >> On 04/ 9/10 01:58 PM, Peter Olcott wrote: >>> >>> Does the MySQL InnoDB storage engine have a journal file >>> like SQLite for crash recovery? >> >> I suggest you do some background reading: >> >> http://dev.mysql.com/doc/refman/5.5/en/innodb.html >> >> To quote the first paragraph: >> >> "InnoDB is a transaction-safe (ACID compliant) storage >> engine for MySQL that has commit, rollback, and >> crash-recovery capabilities to protect user data." > > OK that sounds good. Do you also have to turn off disk write > caching? > I never have. At least on Solaris, disk write caching is always disabled for UFS and trust ZFS to "do the right thing". -- Ian Collins |