*** grantc has joined #ingres | 00:53 | |
*** jundi has joined #ingres | 00:56 | |
jundi | good morning | 00:57 |
---|---|---|
jundi | grantc i was wondering if we can continue the vnode troubleshooting today? | 00:58 |
grantc | hi jundi - sure | 00:58 |
jundi | i was trying to work it out yesterday on my own but no luck.. | 00:59 |
grantc | you're using vb right? are both the server and client machines in virtual box? | 00:59 |
jundi | i am starting to think that it could be related to the virtualbox itself | 00:59 |
jundi | my client machine is the vb host machine | 01:00 |
grantc | and the server is the suse host? | 01:01 |
jundi | the ingres db is installed on the vb guest virtual machine... which opensuse | 01:02 |
jundi | *which is opensuse 10.3 | 01:02 |
grantc | and the client is as well? | 01:02 |
grantc | scratch that | 01:02 |
grantc | what networking are you using for vb - NAT or bridged? | 01:03 |
jundi | no.. the ingres client is installed on the windows box.. my laptop.. which is the host machine for vb | 01:03 |
jundi | bridged | 01:03 |
grantc | so the linux vb machine has an IP on the same subnet as the host? | 01:03 |
jundi | yes | 01:03 |
jundi | and they can ping each other | 01:04 |
grantc | can you ssh in from the host to the virtual machine using putty? | 01:04 |
jundi | no.. i tried that yesterday but it did not work.. | 01:04 |
jundi | the funny thing is.. yesterday i turned off the linux and tried to test the vnode from the netwrok util. it showed the same error.. | 01:05 |
grantc | what does "sudo /etc/init.d/sshd status" | 01:05 |
grantc | show? | 01:05 |
jundi | running | 01:07 |
grantc | so there is your problem or another manifestation of it - Windows -> Linux not working | 01:10 |
grantc | and "sudo /sbin/rcSuSEfirewall2 status" | 01:11 |
jundi | well.. it is running NOW.. i restarted the linux machine a moment ago.. should i turn it of from yast? | 01:14 |
grantc | "sudo chkconfig SuSEfirewall2_init off" and "sudo chkconfig SuSEfirewall2_setup off" will disable it | 01:15 |
jundi | http://ingres.pastebin.com/KrG1kCfy | 01:19 |
jundi | this is the errlog file after i tried to connect after turning off firewall2 as in your instructions | 01:20 |
jundi | the errlog is from the client machine | 01:20 |
grantc | v strange - something else already on those ports? | 01:24 |
grantc | is the windows firewall enabled? are the ports blocked there? | 01:28 |
jundi | no.. nothing.. i used the netstat command to see the used ports on my windows machine.. | 01:28 |
jundi | the windows firewall is disabled. | 01:29 |
grantc | out of interest is there any reason you changed the installation code of Ingres on the client? | 01:30 |
grantc | do you have other installations of ingres on windows? | 01:30 |
jundi | no.. its just that when i first tried to install it.. it gave some errors.. so i just renamed it as trial I2 to avoid any problems or conflicts in the registry.. | 01:31 |
jundi | with the first one.. | 01:32 |
jundi | you know how it is with windows ;) | 01:32 |
grantc | what were the errors from the first installation? | 01:32 |
jundi | missing files and dll's.. i think it was a corrupted download.. i had to download it again.. | 01:33 |
jundi | grantc i am going to try the Host only networking mode for vb | 01:34 |
jundi | This can be used to create a network containing the host and a set of virtual machines, without the need for the host's physical network interface. Instead, a virtual network interface (similar to a loopback interface) is created on the host, providing connectivity among virtual machines and the host | 01:35 |
grantc | not sure that's going to help | 01:35 |
jundi | --- from the vb manual | 01:35 |
grantc | the client installation of ingres cannot open the ports needed | 01:35 |
grantc | can you ssh in to the linux box now? | 01:35 |
jundi | connection refused | 01:36 |
jundi | by the way.. i am using windows 7 on my laptop | 01:37 |
grantc | sounds like you have other problems on your machine | 01:37 |
grantc | if you cannot ssh to the linux box you'll not be able to connect with Ingres | 01:38 |
grantc | I'm not sure about win7 support - you might need to be using Ingres 10 | 01:38 |
jundi | but ingres db is not installed on the windows machine.. its on the linux.. the windows has client only | 01:39 |
grantc | i'm not sure that matters - I know we've been making changes for win7 | 01:40 |
grantc | the fact that you cannot ssh into your VB box is an issue - that should just work, assuming SSH is enabled etc.. | 01:40 |
jundi | well.. i'll figure out how to make ssh work between the two machines and see if this resolves the issue.. | 01:41 |
jundi | i'll update you once that is done | 01:41 |
grantc | does vb support win7? | 01:42 |
jundi | thats what they say in their web site | 01:42 |
jundi | i downloaded the latest | 01:42 |
jundi | 3.1.6 | 01:42 |
grantc | fair enough - i only have it on my mac and i don't use it that often | 01:44 |
*** jundi has quit IRC | 01:59 | |
*** jundi has joined #ingres | 02:15 | |
jundi | grantc: i tried the host only network option on the virtualbox and it worked! | 02:16 |
jundi | now i can ssh and the vnode test is successful | 02:16 |
grantc | so your bridging is broken - that's a new feature in VB iirc | 02:16 |
jundi | so i think it was a problem in the vb itself after all | 02:16 |
jundi | i guess so. | 02:16 |
jundi | well.. any way.. thanks a lot for your help.. | 02:17 |
grantc | no problem - glad you got it working | 02:18 |
*** jundi has quit IRC | 02:19 | |
*** KermitTheFragger has joined #ingres | 03:00 | |
*** atrofast has joined #ingres | 03:01 | |
*** Dejan has joined #ingres | 03:08 | |
Dejan | hello everybody | 03:09 |
Dejan | guys, check this out - http://codepad.org/DV49rDrK | 03:09 |
Dejan | i get bunch of similar "errors" | 03:10 |
Dejan | i am running verifydb as user "ingres" (DBA) | 03:10 |
Dejan | it seems normal to have these problems because these are system catalogs | 03:11 |
*** cthibert has joined #ingres | 04:37 | |
*** rossand has joined #ingres | 04:53 | |
*** ChanServ sets mode: +o rossand | 04:53 | |
*** zxiiro has quit IRC | 05:40 | |
*** zxiiro has joined #ingres | 06:24 | |
*** ChanServ sets mode: +v zxiiro | 06:24 | |
*** atrofast has quit IRC | 07:08 | |
*** atrofast has joined #ingres | 07:12 | |
*** Dejan has quit IRC | 07:41 | |
*** pboro has quit IRC | 07:41 | |
*** Dejan has joined #ingres | 07:50 | |
*** pboro has joined #ingres | 07:50 | |
*** awatkins has joined #ingres | 08:14 | |
awatkins | hello. does anyone know the meaning of IIAPI_ST_NOT_INITIALIZED when using hte ruby driver? | 08:15 |
grantc | you probably don't have II_SYSTEM defined | 08:17 |
awatkins | right, i don't. technically the client machine doesn't have Ingres fully installed. | 08:19 |
awatkins | it had just enought compiled to compile the ruby driver | 08:19 |
awatkins | hmm, now that i'm looking at it, it sounds like i have to have a couple of locally running applications, i can't connect directly | 08:23 |
grantc | unfortunately you need the networking component of Ingres running if you want to connect to a remote instance | 08:24 |
awatkins | grantc: will it matter if hte network components are from the 9.3 32 bit osx release while the ruby adapter is using the 10.0 64bit compilation | 08:29 |
grantc | most likely yes | 08:29 |
grantc | if it were the other way round it might work (assuming the 64 bit build also builds the 32bit files as well) | 08:30 |
awatkins | grantc: in the end, we're actually connecting to a vectorwise install. will this cause problems if i'm using v.10.0 or 9.3? | 08:36 |
grantc | that's a good question - I don't know what release of Ingres VW is based on | 08:37 |
grantc | i suspect 10 but they might have branched and earlier code line | 08:37 |
*** Dejan has quit IRC | 08:39 | |
*** cthibert has left #ingres | 08:54 | |
*** Mud has joined #ingres | 09:16 | |
*** Mud has quit IRC | 09:40 | |
awatkins | grantc: do you know how to tell the build process to only build a 64 bit version? | 09:46 |
awatkins | not the 64+32 hybrid? | 09:46 |
*** Mud has joined #ingres | 09:46 | |
grantc | you may be able to edit src/tools/port/conf/VERS.int_osx and set 64BIT to TRUE | 09:48 |
grantc | DarylM might know | 09:48 |
awatkins | grantc: yeah i did that but in some parts its still trying to build 32bit | 09:49 |
DarylM | IVW is based on the 10 (mainline) code but is currently a branch. The sync of the branch to changes in the "10" mainline are at irregular intervals. | 09:50 |
DarylM | There is no IVW port to Mac at this time | 09:50 |
grantc | sure but awatkins was asking if there was a way to do a 64 bit only build for OSX | 09:51 |
DarylM | ;-) trying to catch up | 09:51 |
DarylM | My build is ONLY 64 bit. There is no hybrid Mac at all. | 09:51 |
awatkins | hehe, yeah, cause the mac is just acting as a client, the main server is 64bit linux, which installed just fine | 09:51 |
awatkins | just trying to ge tthe build to work for connectivity | 09:51 |
DarylM | Client as in Ingres/NET? | 09:52 |
awatkins | DarylM: how'd you do that? I set 64BIT=true like grantc said, but then i got this error while compiling: | 09:52 |
awatkins | no, ruby client | 09:52 |
DarylM | Ruby uses Ingres/Net so yes. :-D | 09:52 |
awatkins | gcc -arch i386 -o /usr/local/src/ingres-10-src/main/install/build/ingres/bin/iimerge | 09:53 |
awatkins | oh sorry | 09:53 |
awatkins | thought you meant the .Net connector | 09:53 |
awatkins | am i doing this completely wrong? Can you just build hte Ingres/Net seperately? | 09:54 |
DarylM | Something I think we would all like to see in the great beyond is for the various language systems (Ruby, PHP, Perl, Python, etc) to use the "DAS" server like the JDBC and .NET drivers. But I am sure grantc will groan over the implication of the amount of work that would take..... | 09:54 |
DarylM | Give me a sec..... | 09:55 |
* grantc won't moan but my boss will | 09:55 | |
DarylM | Still checking. All I have set is 64BIT = "TRUE" ; in the VERS.int_osx file | 09:56 |
grantc | if there is enough interest in doing a C-DAS layer, required for PHP/Python/Ruby, let me know - i'm aware of someone who is looking to make it happen | 09:57 |
awatkins | does C-DAS, mean a direct connection tot he server wihtout going through a the Ingres/Net (which sounds like a proxy) | 09:57 |
DarylM | Wait, I also have | 09:58 |
DarylM | #set: option 64BIT | 09:58 |
grantc | awatkins: yes | 09:58 |
awatkins | grantc: in that case, yeah, I definitely think thats the right way to go | 10:00 |
grantc | the Python/PHP/Ruby drivers are written in C and there is no C interface to connect with the DAS server. .NET and JDBC have a protocol driver written in their native languages. In theory you could do the same in PHP etc but it would be better to go via C | 10:00 |
*** KermitTheFragger has quit IRC | 10:00 | |
grantc | time to knock off for the day | 10:01 |
grantc | catch you all tomorrow | 10:01 |
*** grantc has quit IRC | 10:01 | |
awatkins | DarylM: is the set command also in VERS.int_osx? | 10:02 |
DarylM | checking something... | 10:06 |
DarylM | Interesting, I removed the plain "VERS" file and the system still built. There is too much magic to this process..... | 10:15 |
DarylM | But, I have VERS.int_osx editted with the two critical 64 bit setting lines. | 10:15 |
FrankW | amen brother ... so much magic. | 10:15 |
awatkins | i agree | 10:15 |
*** mercede has joined #ingres | 10:28 | |
mercede | hi | 10:28 |
mercede | is any ingress expert here | 10:28 |
mercede | Im stuck in certain issue and new to ingress, can anyone help? | 10:29 |
mercede | anyone like to help? | 10:29 |
mercede | I guess all of you are just bunch of assholes wasting internet bandwidth ! | 10:30 |
*** mercede has quit IRC | 10:30 | |
FrankW | Well, I'm glad I didn't put myself out for him! | 10:31 |
DarylM | I'm sure he gives his clients 60 second turn around..... | 10:32 |
awatkins | DarylM: to use Ingres/Net, do I have to do the full install of Ingres, or can i just fire up the ingresnet program? | 10:45 |
DarylM | You only need the iigcn process and iigcc (Ingres/NET) process. iisunet sets it up. | 10:59 |
DarylM | But if you install from "ingbuild" there is a networking install option | 10:59 |
awatkins | ok cool, thx | 11:00 |
awatkins | I'm following teh instructions in the readme, from building from source. and with createdbms, its can seem to find ingres.tar: | 11:31 |
awatkins | tar: Error opening archive: Failed to open '/usr/local/src/ingres-10-src/main/release/ingres-10.0.0-00-gpl-apple-darwin-i386/ingres.tar': No such file or directory | 11:31 |
awatkins | is there somewhere its supposed to live to be copied over | 11:31 |
*** cthibert has joined #ingres | 11:40 | |
*** Mud has quit IRC | 11:46 | |
*** atrofast has quit IRC | 11:51 | |
awatkins | ah, nevermind readme_into_osx.html was stopping the tar from being created | 12:06 |
*** withdefault has left #ingres | 12:53 | |
*** awatkins has left #ingres | 12:53 | |
*** cthibert has left #ingres | 13:30 | |
*** zxiiro has quit IRC | 13:35 | |
*** rossand has quit IRC | 13:43 | |
*** zxiiro has joined #ingres | 14:20 | |
*** ChanServ sets mode: +v zxiiro | 14:20 | |
*** awatkins has joined #ingres | 14:30 | |
*** ingres has joined #ingres | 14:40 | |
*** ingres has quit IRC | 14:45 | |
*** HY_ has joined #ingres | 14:45 | |
awatkins | DarylM: do you know anything about IIAPI_ST_NOT_INITIALIZED | 15:13 |
DarylM | sounds like a normal error. The API interface was not initialized. | 15:17 |
DarylM | never run into it myself | 15:17 |
awatkins | DarylM: how do you initialize the API interface? this is being thrown by the ruby driver | 15:19 |
DarylM | Not sure. It should just work. grantc will know more about that one but he is on western europe time. | 15:20 |
awatkins | yeah he initially thought itw as because II_SYSTEM wasn't enabled and then because i wasn't running Ingres/Net, but now thats all taken care of | 15:22 |
DarylM | It appears you did not invoke the "initialize" method on the driver interface | 15:23 |
DarylM | But my memory is vague. I would have to look at some of my ruby examples (which I haven't touched in a while...) | 15:24 |
DarylM | Ah, it should have initialized when the class was instantiated. | 15:25 |
DarylM | print "Basic Ingres Test\n" | 15:25 |
DarylM | require 'Ingres' | 15:25 |
DarylM | c = Ingres.new | 15:25 |
awatkins | yeah, thats hte basic | 15:25 |
awatkins | then this | 15:25 |
awatkins | c.connect_with_credentials( "MYHOST::db","REPORTS","REPORTS" ) | 15:25 |
awatkins | and then immediate error | 15:25 |
DarylM | I just synced my svn repository and compiled and installed using -arch=x86_64 and it works. Hmmmm | 15:31 |
awatkins | compile the ruby driver? | 15:34 |
awatkins | compiled? | 15:34 |
DarylM | Well, yes. I don't have a precompiled one for 64bit Mac OS/X. | 15:39 |
DarylM | So I just built and installed it from the source package | 15:39 |
DarylM | Something changed. I had trouble doing that in the past and this time it compiled and built perfectly and installed no problem. | 15:40 |
DarylM | Try catching grantc tomorrow | 15:41 |
* DarylM hears the dinner call | 15:41 | |
awatkins | ok thx | 15:42 |
awatkins | yeah, i think it might be because of the connection to a VW instance | 15:46 |
awatkins | is anyone still around? | 17:25 |
*** rossand has joined #ingres | 18:21 | |
*** ChanServ sets mode: +o rossand | 18:21 | |
*** cytrinox_ has joined #ingres | 19:20 | |
*** cytrinox has quit IRC | 19:23 | |
*** cytrinox_ is now known as cytrinox | 19:23 | |
*** rossand has quit IRC | 19:28 | |
*** zxiiro has quit IRC | 20:10 | |
*** zxiiro has joined #ingres | 20:11 | |
*** ChanServ sets mode: +v zxiiro | 20:11 | |
*** HY_ has quit IRC | 20:12 | |
*** DarylM has quit IRC | 22:09 | |
*** DarylM has joined #ingres | 23:01 | |
*** Mud has joined #ingres | 23:34 | |
*** grantc has joined #ingres | 23:34 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!