2007-02-21 GnuCash IRC logs

00:26:46 *** Demitar_ has quit IRC
00:26:46 *** Demitar has quit IRC
00:26:52 *** Demitar__ has joined #gnucash
00:26:54 *** Demitar has joined #gnucash
01:11:20 *** dbr has quit IRC
01:18:19 *** hampton is now known as hampton|away
01:20:46 *** Wilddev has quit IRC
02:58:37 *** benoitg has joined #gnucash
02:58:49 *** benoitg has left #gnucash
05:12:45 *** Rol1 has joined #gnucash
06:26:49 *** egsavage has joined #gnucash
06:47:12 *** egsavage has quit IRC
06:48:16 *** egsavage has joined #gnucash
07:13:25 *** twunder has joined #gnucash
07:18:47 *** egsavage has quit IRC
07:19:43 *** twunder has quit IRC
07:47:57 *** IanL has joined #gnucash
08:11:49 *** twunder has joined #gnucash
08:24:20 *** v123 has quit IRC
08:32:20 *** nomeata has joined #gnucash
08:35:47 *** nomeata has quit IRC
08:40:01 *** exper has joined #gnucash
08:46:20 *** exper_ has quit IRC
09:16:30 *** warlord-afk is now known as warlord
09:20:17 *** hampton|away is now known as hampton
09:53:04 <jsled> hampton: you'd mentioned filing a bug re: gtk html and printing? Do you have the number handy?
09:54:07 *** Demitar__ is now known as Demitar_
10:02:18 *** IanL has quit IRC
10:18:27 <warlord> jsled, hampton: So what do we do? make a configure test for gtk_html_print_page() and use that if it exists and use gtk_html_print() otherwise?
10:19:04 <jsled> Or base it on the version, like g_date_set_time{,_t}.
10:20:03 <warlord> jsled: that's not really a good idea, because they removed gtk_html_print() without upping the SO-number..
10:20:11 <hampton> I don't recall filing a bug.
10:20:12 <warlord> So we should swap over to the new API when we find it.
10:20:33 <warlord> ... otherwise someone can change the library from under us and now gnucash will runtime-fail.
10:20:58 <hampton> Was there any overlap of the two functions in a release? If not, we'll have to test for the new one and use it if present.
10:21:12 <jsled> True. Of course, that's Not Our Problem.
10:21:48 <hampton> I don't think we can solve that problem, unless the calls to gtk_html_print* are all linked dynamically
10:22:05 <warlord> I believe there was an overlap.
10:22:55 <hampton> So version n introduced gtk_html_print_page and n+1 removed gtk_html_print? Then we should be able to key it on gtkhtml version number.
10:23:52 <warlord> hampton: I believe so.. But they didn't change the so version at n+1 so one could build against n and then "update" to n+1.
10:24:48 <hampton> Right, but gnucash should be able to look for <n and use gtk_html_print, or >=n and use gtk_html_print_page
10:24:52 <warlord> I can tell you, though, that libgtkhtml 3.10.3 does NOT have gtk_html_print_page
10:26:39 <warlord> Is basing it on the version any better/easier than looking for the function?
10:28:01 <hampton> If my builds are right, and I'm not saying they are, FC[567] all use libgtkhtml-3.8.so.15.3.9
10:29:09 <warlord> what does rpm -q gtkhtml3 give you? On FC5 I've got gtkhtml3-3.10.3-1.fc5.1
10:29:12 <hampton> Depends on how many things changed. If its just the one, then its probably an easier configure test to check the function existence.
10:29:32 <hampton> gtkhtml3-3.12.3-1.fc6
10:29:46 <jsled> I'm at gtkhtml-3.12.2
10:30:15 <warlord> [warlord@cliodev ~]$ pkg-config --modversion libgtkhtml-3.8
10:30:15 <warlord> 3.10.3
10:30:22 <hampton> Those earlier numbers came from my jhbuild trees, so it might not be building the right versions.
10:30:53 <warlord> okay..
10:32:44 <hampton> so with what version did the problem start?
10:34:24 <jsled> That's what I'm trying to find out. I just made gnome viewVC puke trying to figure out which tag/release r8386 corresponds to.
10:34:43 <warlord> hampton: I dont know when it started...
10:34:46 <jsled> Looking at http://ftp.gnome.org/pub/GNOME/sources/gtkhtml/3.0/>, LATEST-IS-3.0.10, which obviously isn't true. :p
10:35:05 <warlord> LOL
10:36:02 <jsled> Okay, http://ftp.gnome.org/pub/GNOME/sources/gtkhtml/3.12/ has 3.12.3 as latest ... and...
10:36:18 <jsled> http://ftp.gnome.org/pub/GNOME/sources/gtkhtml/3.13/ looks like development versions.
10:38:09 <jsled> So, digger vermont was having problems with Feisty, and going between 3.13.6 .. 3.13.91
10:38:19 <hampton> [ot] Wow, its 43F outside and headed to 50F today.
10:38:27 <jsled> or something; at least, that's the diff he'd posted before
10:39:02 <jsled> I wonder if they just haven't incremented the soversion *yet*...
10:39:40 <jsled> I'm going to ask Ota what gtkhtml version is in play, there. It's hard to understand why it'd be the dev branch version.
10:42:56 <hampton> ESR gives up on Fedora ... http://www.linux.com/article.pl?sid=07/02/21/1340237
10:58:22 <warlord> Yeah, well, ESR likes writing letters about every time he takes a poop.
11:01:45 <warlord> And strangely, people still find this interesting.
11:23:42 <hampton> Linus is dissing Gnome... ESR is dissing Fedora... what's next?
11:24:21 <hampton> A good old fist fight?
11:24:28 <hampton> :-)
11:25:35 <jsled> Is there a good pointer to the Linus/Gnome stuff? I've only heard bits
11:26:01 <warlord> Hey, I have lots of issues with Gnome, too.
11:26:09 <warlord> I dont blame him for dissing it.
11:26:47 <hampton> I've just seen his email on the LKML.
11:27:10 <jsled> Well, I guess it depends on what he's complaining about...
11:28:38 <jsled> Nice. http://google.com/search?q=linus+hates+gnome seems to find what I want. :)
11:35:47 *** egsavage has joined #gnucash
11:36:21 <warlord> Hey, at least /I/ always respond to a complainer by saying "send in a patch" :-)
11:37:45 <hampton> bbl
11:37:48 *** hampton is now known as hampton|away
11:38:05 <warlord> see ya hampton|away
12:11:17 *** warlord is now known as warlord-afk
12:14:43 *** prock_ has joined #gnucash
12:23:43 *** prock has quit IRC
12:34:36 *** tedc has joined #gnucash
12:37:47 <tedc> Having trouble getting the help to work w/ 2.0.5. Strace shows that /usr/local/share/gnucash/gnome/help/gnucash/gnucash/C/gnucash-guide.xml is being opened but there is a KDE help center errorThere is no documentation available for /usr/local/share/gnucash/gnome/help/gnucash/gnucash/C/gnucash-help.xml.
12:52:14 <jsled> Is it in /usr/share/[...]? What distro? How'd you install gnucash?
12:55:58 *** Demitar__ has joined #gnucash
12:56:01 <tedc> SuSE 10.1 Kernel 2.6.16.13-4 . compiled and make installed, ditto for the docs.
12:56:50 <jsled> tedc: there's recent traffic on gnucash-user about gnucash-docs on SuSE.
12:57:16 <jsled> Also, <http://wiki.gnucash.org/wiki/GnuCash> has a pointer to OpenSuSE 10.{1,2} packages, which you might find useful.
13:05:20 *** Demitar has quit IRC
13:09:45 *** andi5 has joined #gnucash
13:09:45 *** gncbot sets mode: +o andi5
13:13:00 *** sjc has joined #gnucash
13:13:56 *** warlord-afk is now known as warlord
13:14:17 <andi5> warlord: i guess you know #408898
13:22:19 <warlord> Nope, why should I?
13:30:09 *** ErKa has joined #gnucash
13:51:35 *** ErKa has quit IRC
14:01:18 <andi5> jsled: http://wiki.gnucash.org/wiki/Windows#Report_error indicates that guile stacktraces will not be printed to /tmp/gnucash.trace ... is that right and/or fixable?
14:02:14 <tedc> Too many failed dependencies in the Rauch OpenSuSE10.{1,2} rpm's to troubleshoot. Since guncash compiles, installs and appears to run just fine, I'll look at the SuSE help problem.
14:04:25 <andi5> tedc: i think mr rauch has made a bug report somewhere... but do not ask me :)
14:16:31 *** |gunni| has joined #gnucash
14:17:23 *** hampton|away is now known as hampton
14:25:41 *** Humanoid has joined #gnucash
14:33:40 <Humanoid> Is there any way to see the time of a transfer, instead of just the date?
14:36:13 <warlord> Humanoid: http://bugzilla.gnome.org/show_bug.cgi?id=89439
14:38:36 <Humanoid> warlord: Wow, the bug was opened on 2002-07-30, and it still wasn't fixed.
14:41:53 <Humanoid> I transfered money from one bank to another bank. I've imported the OFX files from each bank, and they both have the transaction, except that it's at a different time. Can I tell gnucash to _connect_ the two transactions to each other, even though the debit and credit are done at slightly different times in the two accounts?
14:44:48 <warlord> Humanoid: it's not a bug. It's a "Request for Enhancement"
14:45:17 <warlord> Humanoid: you need to have marked them as duplicates during the import.
14:49:04 <Humanoid> warlord: How do I mark it as a duplicate? I see two boxes, one for "A" and another for "R", but I can't find documentation on their meanings.
14:50:01 <jsled> andi5: Good question ... I'll take a look.
14:51:38 <Humanoid> I clicked on "R", and then it said "match missing" on the right, so then I clicked on "match missing", and it opened a dialog with the bottom half saying "Potential splits matching the selected transaction:", but none are listed.
14:52:15 <warlord> Humanoid: how far off are the dates? Amounts? Did you mark the same cross accounts?
14:52:24 * warlord has never used the OFX importer.
14:52:40 <Humanoid> The withdrawal from the first account is at noon, and the deposit into the other bank account is at 2 am the next day.
14:52:56 <Humanoid> The amounts are exactly the same.
14:53:18 <Humanoid> I don't understand the third question: "Did you mark the same cross accounts?"
14:54:18 <warlord> transactions have (at least) two accounts, the "source" and the "destination"..
14:54:41 <warlord> With OFX the source is already known, but did you properly set the "destination"?
14:54:42 <jsled> andi5: huh. src/app-utils/gfec.c:error_handler calls g_warning(...); given the default log-domain for app-utils is 'gnc.app-util', well, that's where the message should go. I'll need to dig deeper, at some future point. I'll make a note.
14:55:13 <andi5> g_warning sounds.. low
14:55:25 <Humanoid> warlord: No, I didn't do that when I imported the first set, I'll try that now before importing the second file.
14:55:30 <warlord> jsled: I wonder if there's some way to set the log domain in the guile environment?
14:55:32 <jsled> They should probably be g_critical.
14:56:30 <andi5> jsled: thanks
14:56:39 <jsled> warlord: not really right now; the gnc:{message,error,warn,debug} methods now call gnc-scm-log-$1, with an explicit "gnc.scm" domain. I guess there could be variant versions that take the domain as a param...
14:56:55 <jsled> s/with an/which uses an/
14:57:51 *** andi5 has quit IRC
14:59:44 <Humanoid> warlord: Okay, after having set the "transfer to" account, when I imported the second file from the second bank, it did a "reconcile auto match". But all that did was to ignore the new import, throwing away all the info from the second import.
15:00:36 <warlord> Which is mostly correct...
15:00:56 <warlord> (it should have added a second OFX TXN-ID, but that might not be possible given the current implementation)
15:01:23 <Humanoid> I guess I'll just create a "limbo" account, and when I transfer from one bank to the other, I'll have take the withdrawal at noon go to the limbo account, and the deposit at 2 am the next day is taken from the limbo account.
15:06:12 *** Demitar__ has quit IRC
15:12:09 *** Humanoid has quit IRC
15:12:39 *** Demitar__ has joined #gnucash
15:28:29 *** Esaj has joined #gnucash
15:40:47 *** tedc has quit IRC
16:12:25 *** MrN has joined #gnucash
16:12:55 <MrN> hi
16:15:56 <exper> SaLuT A tOuS !!!!!!!!!!!!!!!!!!!!!!!!!!!
16:16:43 <exper> bon personne d'interessant j'me casse
16:17:02 <warlord> Excuse me?
16:19:47 *** Esaj has quit IRC
16:20:29 *** Anthony_ has joined #gnucash
16:20:30 *** Anthony_ is now known as Anthony
16:21:18 <Anthony> Hello...I'm a newbie trying to get GnuCash to import an old Microsoft Money back up file, but no luck on SuSE 10.2...any thoughts?
16:22:37 <jsled> GnuCash cannot import Microsoft Money files.
16:23:36 <Anthony> uhh...the package description said that it could -along with Quicken, etc.
16:23:46 <warlord> Anthony: you need to use MM to convert the file to QIF and then import the QIF.
16:23:58 <jsled> The description where?
16:24:21 <Anthony> warlard: okay, thank you.
16:24:49 <Anthony> jsled: The description was in the SuSE 10.2 YaST Software Management section.
16:27:02 <jsled> Is there a package maintainer listed?
16:27:19 <Anthony> none that I could see
16:28:18 <cortana> listing the maintainer of the package would be far too confusing for the poor users! ;)
16:30:34 *** ErKa has joined #gnucash
16:38:29 *** tedc has joined #gnucash
16:45:50 *** motin has joined #gnucash
16:48:42 *** Anthony has quit IRC
16:55:50 *** ErKa has quit IRC
17:02:46 *** twunder has quit IRC
17:03:08 *** ErKa has joined #gnucash
17:19:27 <hampton> Looks like we need to ban gtkhtml3 >= 3.16.6
17:19:50 <warlord> It does.. But there's no way to ban it at runtime.
17:20:34 <hampton> There's no way to get the library version number? Can we read the file name?
17:21:55 <warlord> at runtime?
17:22:30 <hampton> yes. iirc glib and gtk both provide functions to get version information
17:22:48 <hampton> I'm wondering if gtkhtml3 does too
17:24:20 *** bonez39 has joined #gnucash
17:25:55 <jsled> I wonder if we can encourage them to increment their so number.
17:26:16 *** twunder has joined #gnucash
17:26:36 <hampton> That would certainly be the right thing for them to do.
17:27:09 <hampton> And the right solution to everyone's problem.
17:27:09 <jsled> I find it funny, since many recent releases of gtkhtml have seemed to increment the SO number for no good reason, but this one doesn't.
17:27:41 <hampton> I just wish they'd fix their pkg-config file
17:29:37 <warlord> Well, Bill reported a bug about them not increasing the so-number
17:30:23 *** egsavage has quit IRC
17:30:39 <jsled> Ah, I see.
17:31:26 <warlord> http://bugzilla.gnome.org/show_bug.cgi?id=408671
17:34:12 *** egsavage has joined #gnucash
17:39:33 *** tedc has left #gnucash
17:44:13 *** egsavage has quit IRC
17:45:19 *** franz has quit IRC
17:47:12 *** franz has joined #gnucash
17:59:06 *** nomeata has joined #gnucash
17:59:12 *** egsavage has joined #gnucash
18:01:37 *** hampton is now known as hampton|away
18:42:30 *** twunder has quit IRC
18:57:29 *** tedc has joined #gnucash
18:57:53 *** warlord has quit IRC
19:08:15 *** motin has quit IRC
19:08:20 <MrN> n8
19:08:40 *** MrN has quit IRC
19:15:57 *** Rol1 has quit IRC
19:20:01 *** warlord has joined #gnucash
19:20:01 *** gncbot sets mode: +o warlord
19:21:29 *** Rolf has joined #gnucash
19:21:47 *** |gunni| has quit IRC
19:26:26 *** nomeata has quit IRC
19:31:47 <tedc> To fix the documentation in SuSE cd /opt/gnome/share/gnome/help; cp -rpv /usr/local/share/gnucash/gnome/help/gnucash .
19:32:09 *** exper_ has joined #gnucash
19:34:04 <warlord> tedc: why not just cd /opt/gnome/share/gnome/help ; ln -s /usr/local/share/gnucash/gnome/help/gnucash .
19:34:44 <tedc> certainly
19:39:41 *** exper has quit IRC
19:41:46 *** jpeach has joined #gnucash
19:42:11 <warlord> If the LN works you should add it to the SuSE wiki page..
19:42:12 *** jpeach has left #gnucash
20:02:52 <jsled> Hmm. Did an updated ChangeLog get generated for the 2.0.5 release?
20:05:56 <jsled> No, it's there.
20:06:43 <jsled> Oh, I see. It should probably be generated with `svn log -v`, to get the changed-file list.
20:07:08 <jsled> warlord: I'm going to assume you're already replying to TB?
20:07:15 <warlord> Yes.
20:07:28 <warlord> Not for 2.0.x
20:07:37 <warlord> for 2.2.x we're going to use svn log..
20:07:45 <jsled> I see.
20:08:26 <warlord> I thoguht that cstim had created some automated procedure for it..
20:08:36 <jsled> Yeah, it does appear that it has '-v'.
20:08:50 * jsled tries it...
20:09:36 <chris> lots of love in the air tonight. must be springtime somewhere.
20:09:42 <jsled> heh
20:09:45 <warlord> LOL
20:10:20 <warlord> TB is just a little fussy because I pinged him about getting 2.0.5 into Debian and the debian maintainers were asking him about specific changes from 2.0.2 and he either didn't know about trac or didn't care to use it.
20:11:48 <jsled> Well, I think the point is well taken; the distributed software should be self contained, and a condition of the GPL is the enumeration of changes.
20:12:10 <warlord> we did enumerate the changes..
20:12:33 <warlord> just perhaps not as explicitly as the GNU Coding Standards dictate.
20:12:41 *** Wilddev has joined #gnucash
20:12:41 *** gncbot sets mode: +o Wilddev
20:12:52 <warlord> hiya Wilddev
20:12:55 <jsled> true.
20:13:25 <Wilddev> hiya warlord
20:14:32 <warlord> how's it going?
20:15:12 <Wilddev> pretty well, u?
20:16:03 <warlord> Doing okay...
20:18:35 <warlord> Hmm.. That seems to look somewhat different than what cstim had..
20:19:51 <jsled> I think I'd rather have the files listed one-file-per-line (just literally from `svn -v log` rather than comma-fied like ChangeLog format, but I don't care.
20:20:55 <warlord> I dont particularly care about that.. One per line is fine with me. Comma sep'd is fine with me, too. I think a vertical list is probably easier to read.
20:26:21 <jsled> TB's tone is really atrocious.
20:26:38 <warlord> I guess I'm just used to him.
20:40:01 <jsled> Well, there's being terse. But he's just being rude.
20:40:56 <Wilddev> jsled: who's TB?
20:41:17 <jsled> Thomas Bushnell ... the debian maintainer.
20:41:25 <jsled> re: the recent ChangeLog thread on -devel.
20:42:03 <Wilddev> oh right, I noticed that about him before
20:48:06 <warlord> He's a very bright guy, even if a little opinionated at times. ;)
20:50:09 <jsled> Oh, I'm sure he is. And I respect his contributions. But he could have defaulted towards Hanlon's Razor instead of being confrontational.
20:50:48 <warlord> Yeah...
20:51:07 <warlord> Okay, gotta run for a bit.. I'll be back in about 20 minutes.
20:54:21 *** sjc has quit IRC
21:12:34 *** twunder has joined #gnucash
21:18:18 <warlord> back
21:25:03 <warlord> Oh by the way.... Happy Birthday, Wilddev
21:25:18 <Wilddev> :-)
21:25:37 *** twunder has quit IRC
21:25:40 <jsled> Happy Birthday indeed!
21:26:57 <Wilddev> thx jsled
21:28:08 *** franz has quit IRC
21:41:52 <warlord> jsled: with r15645 do you still need the GncGMapFunc cast?
21:45:11 <jsled> Probably not strictly, but I like it.
21:45:15 <jsled> (as documentation)
21:51:03 <warlord> you could use a comment for documentation.. With a cast you could potentially get argument errors ;)
21:51:09 <jsled> Of course, it'd be even more clear if the _reorder_... args were typed correctly.
21:51:21 <warlord> well, there's that too..
21:51:30 <warlord> (gpointer considered harmful?)
21:51:51 <jsled> gpointer considered meaningless, anyways. :)
21:52:53 <warlord> Well, I think it's harmful because of the 'cast' issues..
21:53:10 <warlord> Unless you can use the gtype casts to get a runtime cast error.
21:53:10 <jsled> I don't understand the cast issue... What do you mean?
21:54:39 <warlord> well, casts in general are bad (IMHO)
21:55:08 <warlord> casting a function pointer potentially hides issues that could be caught at compile-time
21:56:21 <warlord> (at least if the function arguments are real types and not gpointers)
21:56:54 <jsled> I see.
22:01:13 <warlord> and casting types is bad for the same reason (unless you use a runtime type-checking system such as dynamic_cast<>() in C++ or the gobject/gtype casts.
22:01:15 <warlord> )
22:03:08 <elb> amen
22:03:39 <elb> Glib/Gtk+ have sort of taught people to just cast everything and not think about why they're casting, unfortunately
22:04:16 <warlord> elb: well, Gtk has a runtime type checking system...
22:04:26 <warlord> (at least for gobject-derived objects)
22:10:51 <Wilddev> cool my raid array is rebuilding
22:11:04 <Wilddev> finally a stable set of disks
22:11:21 <jsled> What kind of disks?
22:12:54 <warlord> which kind of raid?
22:13:11 <Wilddev> ide on raid1
22:13:32 <Wilddev> I bought a couple of seagate 500GB disks and both had problems
22:13:53 <Wilddev> rmaed them and now have got 2 that work without bad blocks
22:13:58 <warlord> Ahh, raid1.. *nods*
22:14:09 <warlord> I'm now debating if I want to do a RAID10 or RAID5..
22:14:31 <Wilddev> you saw the /. article on raid and disks?
22:14:32 <warlord> The more I read, the more I think RAID10 is "better" in terms of data integrity.
22:14:51 <warlord> I'm not sure I saw it on /. per se.
22:15:50 <Wilddev> there was a large study done on disk failures in the real world
22:16:02 *** hampton|away is now known as hampton
22:16:08 <Wilddev> and apparently raid5 makes it more likely 2 disks will fail in an array
22:16:56 <warlord> Yeah, I read that.
22:17:28 <Wilddev> I've been happy with my raid 1 setup the last 6 years
22:18:03 <warlord> the downside of RAID10 is that you only get half your space usable, whereas with raid5 you can get much more.
22:18:14 <Wilddev> true
22:18:40 <Wilddev> I was running raid 10 till recently, 0 on my hardware raid and 1 on software raid
22:20:03 <warlord> For example, assume 8 500GB disks. With raid10 you get a 2TB array. With raid5 (and two hot spares) you get... 2.5TB. You get 3TB with only 1 hot spare.
22:20:18 <warlord> (If I'm doing my math right)
22:22:16 <hampton> happy birthday Wilddev
22:22:23 <Wilddev> thx hampton
22:22:44 <warlord> Wow... 750GB SATA for $310
22:23:04 <jsled> I really wanted a raid setup with my last pair of drives, but the crappy on-board IDE controller just didn't work. When I upgraded my machine in the spring, I upgraded everything except the disk subsystem...
22:23:09 * hampton thinks its time to update his raid5 array
22:23:26 <jsled> ...so this spring/summer I think I'm going to go to a nice SATA array of ... something. Not quite sure, yet. :)
22:24:21 <warlord> I'm still waiting for the quadcore processors.
22:24:45 <warlord> Then I plan to build an 8-CPU VMware host system, and maybe set up some gnucash build environments :)
22:28:09 *** twunder has joined #gnucash
22:30:12 *** exper has joined #gnucash
22:35:39 *** exper_ has quit IRC
22:36:12 *** Wilddev has quit IRC
22:42:57 <elb> warlord: yes, Gtk itself is pretty rigorous -- but many of its callbacks etc. take gpointers, and people use them liberally -- as with the glib data structures
22:43:35 <elb> constructing a gobject type is nontrivial enough that most applications don't bother to do it for every datatype they wish to pass through such generic interfaces
22:46:33 <warlord> elb: True.. And gnucash is no different.
22:47:01 <warlord> oh, w.r.t. RAIDs, I just found this page: http://www.pcguide.com/ref/hdd/perf/raid/levels/comp-c.html
22:49:05 <elb> when you guys are saying RAID10, do you mean RAID0/1 ?
22:49:50 <elb> (that is, stripes across mirrored pairs)
22:50:10 <elb> it's been so long since I looked at disks that I fear my terminology is old ;-)
22:51:05 <elb> it's interesting that they call RAID5's storage capacity N-1/N; I was under the impressoin it was N-k/N
22:51:25 <warlord> Well, there is RAID10 and RAID01.. I prefer the former, which is where you mirror each disk first (RAID1), and then you stripe them to combine the RAID1 mirrors (the RAID0)..
22:52:22 <elb> yes, that's what I figured
22:53:40 <warlord> From my research this sounds like the best combination because when you lose a disk you only have to rebuild that raid1 mirror instead of rebuilding the WHOLE array.
23:03:25 *** twunder has quit IRC
23:26:22 <warlord> okay, bedtime for me. good nigh.
23:26:29 *** warlord is now known as warlord-afk
23:36:43 *** egsavage has quit IRC
23:47:54 *** dbr has joined #gnucash
23:50:59 <dbr> There's definitely a discontinuity between trunk and 2.0.x in the way the generic import transaction matcher does its final handoff to the register.
23:52:11 <dbr> On trunk at least some transactions (from an ofx stream) will be left in the register even if you tell the matcher not to import them.
23:52:28 <dbr> I'll see if I can create a sample file that always fails for me.