As a follow-on to my first post about Amazon Lightsail, here’s some additional information that I’ve discovered…
I ran into an interesting problem with my new server the other day.
Due to a hardware problem with an external USB hard drive, I had to reboot the XEN host.
I had to do the reboot remotely … so I couldn’t intervene if something went wrong. Yes, you guessed it, something did go wrong.
When I finally got home I found the system trying to boot ‘bree’ … which is one of the XEN guest machines.
Well, I’m pretty pleased with CentOS so far.
I just completed a minor version upgrade (5.5 to 5.6) via yum and it went very smoothly. This involved 3 virtual guest and one host (the guests run inside the host).
I had one small glitch with the RPM database on middle-earth. Some dependencies were screwed up, but that was corrected by running
yum clean all.
In order to speed things up, I’m going to be mirroring the most current CentOS package repository on middle-earth … that way I can get the updates done much quicker (installs too).
Ginny has been bugging me about the problem she’s been having posting to her blog from the iPhone and other third party tools that she uses.
The problem has been narrowed down to a bug in LIBXML2 that causes leading angle brackets “<” to be stripped off of posts when processed through the XMLRPC support in PHP. I logged the issue in the WordPress bug tracker, and the consensus is that the problem started manifesting after an automatic update applied by Fedora 8.
Well, to solve the problem I’ve moved Ginny’s blog to a temporary server running in a base Fedora 8 install in a VMWare instance. Luckily WordPress stores most of it’s content in the MySQL database (not the images, unfortunately). A bit of NFS magic to mount the appropriate directory in the virtual system, and it’s working fine. The server itself only has Apache, PHP, MySQL client, SSH, and a few other bits & pieces to keep it running. No automatic updates, GUI, compilers, etc.
Once the bug is fixed (either WordPress, using a different XML processing mechanism, or LIBXML2) I’ll move her server back to the main server.
In the mean time, her blog is going to run in the virtual machine ‘Hobbittown’ (named because it’s small).
Yesterday I found that the RPM database on my Fedora Core 6 linux system’s were corrupted and that the regularly running update process was failing (without telling me, unfortunately).
After fixing the RPM database problem (
rm -f /var/lib/rpm/__db.* && rpm -vv --rebuilddb) and running the update (yum update), I found that SpamAssassin’s update process wasn’t working anymore.
root@rivendell ~]# sa-update
Use of uninitialized value in concatenation (.) or string at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/Scalar/Util.pm line 30.
Apparently one of the updates that were applied in the mass update caused SpamAssassin to break.
The same problem occurred when I tried to test the SpamAssassin rules.
root@rivendell ~]# spamassassin --lint
Use of uninitialized value in concatenation (.) or string at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/Scalar/Util.pm line 30.
A bit of research turned up this link.
Luckily the fix was fairly easy … just update the Scalar-List-Utils CPAN package …
perl -MCPAN -e 'install "G/GB/GBARR/Scalar-List-Utils-1.18.tar.gz"
… and everything worked fine again.
There were a few minor glitches … but nothing that couldn’t be handled.
One thing that was kind of annoying is the fact that the Apache config (httpd.conf) was replaced on both servers (the original was backed up to a ‘rpmsave’ file). This, in itself, wasn’t that big a deal … but I wasn’t expecting it.
I have an idea for a good open source project … that could be incorporated into new distribution releases … an “Upgrade Impact Analysis” tool. It would evaluate your existing configuration, compare to what is known about a new distribution, and tell you what config files could be used without modification and which files would have to be reworked.
Another problem I had, which had me worried, was the fact that I couldn’t get ClamAV to rebuild. After a bit of research, I found that my linker configuration file ‘ld.so.conf’ had a ‘/usr/i486-linux-libc5/lib’ listed. This has to be left over from one of my older Redhat installs. The files in that directory are timestamped April 15th 1999. To be honest, I’m shocked I was able to build anything with that old code in the linkers configuration. Luckily, simply commenting that line out fixed the problem.
So far everything seems to be working OK, so I’m cautiously optimistic.
(with apologies to Lemony Snicket)
Last week was not all that great for me computer wise.
I had three events that pretty much threw me for a loop. If it hadn’t been for Ginny, I would have probably been a drooling, babbling, idiot.
Where should I start? Hmmm, the beginning, maybe?
(cue fade to memory music)
It all started the week before last … I was figuring out what to do with the week off between Christmas & new years. My linux servers (Gondor and Rivendell) are both running Fedora Core 4, and Core 6 was released a little while ago. I figured it would be a fine time to upgrade both servers.
The original plan was to upgrade the servers xmas eve, while Ginny was at church. That plan got changed, however, because Ginny didn’t want her churches website down over Christmas. Ok, later in the week would be fine.
So on Christmas eve, we went to my brother Mitch’s house for dinner. While there, Mitch showed me his Hauppauge MediaMVP
Christmas comes and goes … everything is nice, had a fun time we the family.
On December 26th, my friend Steve calls and asks if I want to take a trip down to Frys to see if they have any deals or specials going on. Sure I say, that could be fun. I’ve been wanting to get the wire to connect my Dell PE430SC server’s AUX_LED to the PCI IDE controller … just so I can see disk activity.
Since I wanted Steve to be able to help me find this cable, I opened the case to take pictures of the connectors. Got my pictures and proceeded to close the case. This is when the trouble happened. I accidentally bumped the ArcoIDE Duplidisk mirroring adapter. This must have caused a temporary problem because the unit started beeping as if one of the drives had failed. A drive hadn’t failed … but the drives were no longer in sync. Well, this shot the trip to Frys.
I spent the next couple of hours trying to figure out what what happened … and get the drives back in sync. I had to call Arco because there were problems getting the adapter recognized by the management software. I ended up attaching the primary hard drive of the mirrored pair directly to the ide controller. Rebuilding the mirror would have to wait a few days.
I decided to fix the mirror and do the Linux upgrade on Saturday the 30th.
I woke up bright and early on Saturday, fed the cat, fed myself, and went down to start my work.
First thing to do is get the mirror re-established. It took a bit of work, because I still couldn’t get the software to recognize the adapter. I determined that it was because the adapter was attached to a add-on PCI IDE card. When I hooked the adapter up to the motherboard’s IDE (it only has one), it worked fine. This is going to take 3 hours for the mirror to be rebuilt. It has to duplicate all the contents of the primary drive to the mirror drive.
While this is running, I started to do the upgrade on Gondor.
While the servers were down, I had setup a dummy “This site is under construction” page on another machine (shadowfax, for what it’s worth). My plan was to assign all the IP’s I have to that one machine so nobody would think my sites were just gone.
Got it all setup … but I couldn’t hit the dummy page. I figured it was because the DSL router still had the old machine’s mac addresses in memory. This has happened before. No problem, I telnet into the router and select the “Restart device” option. The device restarts … but then the 4 lights that correspond to the ethernet ports on the router start blinking at the same time … and the DSL light is not on. I try telneting to the router again, but get no response. I try using the RS232 connection to log into the router, but get no response. Nothing is working.
I also tried connecting my laptop to one of the ethernet ports on the router and notice that windows alternates between saying that my ethernet is Connected and The cable is unplugged at the same frequency as the blinking lights.
I call Covad and talk to a tech … it’s pretty quickly determined that the router had died. The earliest they can get someone out to replace the router is Monday the 1st.
Well, the midrange.com activity will be light because of the holiday … so I shouldn’t expect too many panic calls because there isn’t any list traffic.
So Sunday I decide to get the MediaMVP unit setup.
I had ordered it from amazon.com earlier in the week, but hadn’t had a chance to set it up. My original plan was to use my D-Link DWL-G730AP Wireless Pocket Router
Unfortunately, I couldn’t get the two devices to play together. Well, my boss had given me a $30 gift card for BestBuy, so I figured getting a wireless gaming adapter would be a fine use for it. I went to the Schaumburg BestBuy, but they didn’t have any wireless gaming adapters in stock at all. I then went to the Arlington Heights BestBuy and found they had one. The Linksys WGA54AG Game Adapter 802.11A/G
I got the unit home and started to configure it … but then found out that it didn’t support WPA encryption, just WEP. That’s stupid. Half the reason to use 802.11G is to have better encryption.
I bit of searching online and I found that CDW had the D-Link DWL-G820 Wireless Gaming Adapter, 802.11g, 108Mbps
So off I go to Vernon Hills. I return the Linksys device and go to CDW. Unfortunately, although CDW has the DWL-G820 on site, it’s not off the truck yet. So I have to wait … about 2 hours.
I finally get the new gaming adapter and get it home … it works, although I still haven’t gotten the MediaMVP to work correctly yet. I need to futz with it a bit more.
HAPPY NEW YEAR!Ok, New Years Eve is over … New Years Day is here. The Covad guy is supposed to be here between 8am and noon. Around 10:45 he calls to say he’s in the city, but on his way out … he should be there a little before noon. That’s refreshing.
He gets here exactly when he said he would … runs a few tests, says “Yeah, the router is fried”. Gets the replacement out and starts to configure it. About 30 minutes later, it’s all configured and we’re back on the air. All I have to do is re-implement my firewall filters. Sure wish I had backed them up a while ago. Oh well, they aren’t that complicated … it’s just a bunch of typing.
It takes me about 2 hours to get the router reconfigured the way it should … then I can get the networking on the linux boxes working again.
All in all, I did a lot of work … but ended up right were I started.
Seems to me that this kind of thing has happened before … a few days off from work, a grand plan to upgrade a system or two, and a bunch of failures that cause nothing but problems.
I’m cursed, I tell you, cursed!
[tags]Linux, MediaMVP, Fedora Core, DLink, Linksys, Covad, DSL, Upgrades, RAID, Wireless[/tags]
I had the oddest thing happen just now … I called a credit card company and asked to be removed from their promotional mailing of “Convenience Checks” … and was.
Convenience Checks are the checks they mail you that let’s you get cash from your credit card … and have a service fee attached in addition to the normal finance charge (which kicks in immediately, not after the normal 25 day grace period).
I’ve been getting a huge number of them from Bank of America for my Linux Fund Mastercard. Well, I finally got tired of it and called the issuing bank.
I talked to William, who was quite helpful … I explained that I would never use the check and I just end up shredding them.
William put me on hold for a bit then came back and told me that he had removed me from the mailing lists and added me to the “Paper Suppression” list, which should totally eliminate the junk mail I get from them.
All in all, a very satisfactory outcome. I was pleasantly surprised.
[tags]Credit Card, mbna, Bank of America, Linux Fund, Junk Mail[/tags]
The new tape drive arrived today … I was a bit worried that it would ever show up, as the vendor I purchased from seemed to be pretty disorganized.
It seems to be ok. It’s even new. I was half expecting the unit to be refurbished, which I would be OK with, because the price I paid was $200 less then the next higher vendor.
The only problem with the tape drive is that it’s noisy as hell … louder than both of the servers put together.
One oddity I found … the old tape drive would erase tapes in under a minute … but the new tape drive wants to write over the entire length of the tape. When I issue the
mt erase command, it spins for a very long time.
Anyone want a marginally working SCSI DDS3 tape drive? 😉
[tags]linux, backup, tape, dds3, usb, scsi[/tags]