Back on November 25h, 2013: I started writing this post:
According to this blog, I received this PowerSource 400 on January 21st, 2008 - click here
In answer to the question at the end of that post, the answer was yes. There are 5 other PowerSource 400's in service in my home.
Anyways...back on October 28th (my 45th birthday ), about an hour after getting into bed there was a power blink....I recall being woken by the PowerSource 400 in the bedroom, but I was still able to breath, so I eventually went back to sleep. I woke in the morning to find that half my network had been down for quite some time, though I didn't see any alerts on the security system control pad.
This being the "network closet" UPS...it was providing backup power for my DSL modem, its associated router, the switch that connects the various rooms in my home together, namely the "living room" and the "back bedroom". Not sure which AP the "side bedroom" was connected to. Though the power outage knocked out my laser printer, which I haven't felt the urge to find out why. Really need to work on finding a more accessible location for it, or work on reversing the entropy in my nuclear room(s)... especially since my other printer is circling the drain, and not sure when I'll get around to getting the new inkjet printer that I've been eyeing.
Restarting the UPS, was when I finally got alerts from my security system. It was a combination of events that happened when I cycled power and events that had been waiting for the return of connectivity. The security system's primary connection is broadband, but it also has a phone line for backup (there's a cellular option, but its been out of stock whenever I've felt the urge to splurge for it....though I kind of have cellular backup anyways....)
It was the return of power that caused it to report loss and recovery of the phone line, along with the queued messages of broadband and power outages, with restores of all... Its built in backup battery had kept it up during the outage (approx 8.5 hours). The phone line from the base station is connected to an ObiHai OBi110. Which has an FXO port to allow selection of analog phone service, or in the event an outage pass through to this. In normal operation the OBi110 is making my Google Voice available to my regular phones, where the OBI110 is configured to use my other Internet gateway. The FXO port is connected to a Cobra PhoneLynx Bluetooth Cell to Home Phone Adapter which is associated with a phone with AT&T prepaid service.
Though I plan to at a later date replace both the OBi110 and PhoneLynx and cellphone with an At&t Wireless Home Phone Base. Which I have sitting, waiting for when I get around to it.
The plan is the OBi110 will move to the living room for use there, though its in associating with the planned inkjet printer replacement.
So, its time to replace the batteries...guess I'll do research on where I can get a pair. Hmmm, it uses the same kind as the pair I had replaced in my SmartUPS 1400 back on October 10th (it went around October 5th, and I ordered them on the 6th). The site provides discounts if you buy multiples. I got 5% off for getting 2, would've gotten 7% off for getting 4, or 8% off for getting 6 (9% for 8 and 10% for 10+).
I hesitate, since I'm waiting to see how the recent problem with BillMeLater goes...so needs to be sometime after Nov 5th to order. Meanwhile a storm comes in on November 3rd, and power is blinking like crazy...and I'm constantly restarting the PowerSource 400. Finally I take it out of service, and everything is just in a regular surge strip. Things remain stable (dbox stayed up straight) until this morning. But, feel that I need to put the PowerSource 400 back into service before my annual Thanksgiving trip.
After much procrastination, and lots of searching. I settle on ordering 4 such batteries from an eBay seller on November 14th. Plan to replace the batteries in my second PowerSource 400 soon, since it was purchased on December 31st, 2008 and provides backup power for my CPAP. (see here).
And finally, this morning, I decided that I've put it off too long and to replace it this morning.
Which was a challenge....
There was supposed to be a page 3, but all I had typed was the page break...so had to figure out how to open it up again. Aside from the 4 corner screws, I remembered that there's a screw behind the sticker in the middle. But, I forgot the rivets in the bottom, and that both parts need to be extracted.
Once opened, I started to disconnect the batteries, and once again threw sparks and added another mark to my screwdriver. Taking a step back, I decided there's an order of operation involved here. The answer is start with the outside connections, and then the inside ones. And, then reverse it. This time I knew the side panels interlock with the batteries to help lock them into place, so made sure the wires were clear, I noted that the neural wire for the front outlet had a kink, I made sure that all the wires were in the channel while buttoning things back up.
Surrounding this, was the challenge of getting it out from behind my headboard, though while I was down there, I found a couple missing packages of mask diffusers, I knew I wasn't using them up that quickly, but couldn't figure out why I kept running out. There was one Xyrem pill bottle, I had one night knocked both down...luckily while the empty (first dose) one went all the way down, the other was just wedged between mattress and headboard in upright position. Since I use bottles longer than a month, I usually go by how grungy the caps get...since I customize the bottles, so I can tell the two apart by touch, which matters when one has a larger dose than the other....and the caps get grunge depending on how many nights I overfill (have always had a problem measuring drinks that way.... ) Or how often they get tipped...
Pages: 1· 2
So, near the end of July, I started investigating (once again) on replacing my HP Photosmart 8450xi (which was now over 8 years old....bought it on June 30th, 2005 - Back from Vacation Tech Buying Spree?...setup on July 9th, 2005 - link
I had started looking some time before this, but was put off for a bit due to my experiences with the Brother DCP-7065DN -- link, since it seemed most of the choices out there were GDI and I'm moving to more and more heavily FreeBSD as my primary operating system.
Especially since it appears that 'box' finally called it quits on December 2nd, before I had started my journey home from Chicago TARDIS that day....and orac is inching close to its end, as the pair of ST2000DL003's which evidently only had 1 year warranties from June/September 2012 started going shortly into the new year. I was trying to use ddrescue to force sector remapping on the first drive, when the other drive has decided to vanish permanently. I had thought it was was DM's that had 1 year and DL's that had 5 years, perhaps I had it backwards....or its a question of when I purchased them, or how they were packaged.
Checking my order history, I purchased one drive on June as a bare drive and later in September as a retail kit. I haven't yet pulled the drives, so I can't look up the serial number for the vanished one, but Seagate's website says the one that is responding is out of warranty. Even if the other drive is still under warranty, not sure I want to deal with getting it exchanged for a refurb to create a solo 2TB drive. Can't think of not wanting raid given what I'll likely use it for. And, not sure I'd buy a different 2TB drive to be its mate (and it won't work with my other 2TB arrays, since its an advanced format 2TB drive...while the lraidz2 pool on zen used legacy format 2TB drives (which limits options of growing it non-destructively.)
Fortunately, I had copied one of the big volumes from it over to zen (along they way it got corrupted, so had been trying to copy it back from zen when the other drive died). And, files of the other volume (my pyTiVo store) should all be in backup, where I don't have space on zen to restore them yet.... I have pyTiVo on zen, but the content under it is different...and larger, so much that it is currently not being backed up. I haven't gotten made much progress on building the second backup server....guess I'll need to look at this sooner than later.
And, now it seems the other 2TB RAID-1 array on orac is dying. I just went ahead and failed the drive that was giving it issues. Not sure what to do with it...suppose I could try ddrescue on it and see what happens. The big volume on it had also been copied over to zen, so guess I'll update my HSTi's to point to zen instead of orac for their content. Another used to be for Time Machine backups, but I had moved that over to zen when I set up the new work laptop to do Time Machine backups on my home network. I was using that space as overflow from pyTiVo. And, another was for backups of various things, which I had stopped adding to as new backups are going to zen now. Its things like regular backups of my websites at dreamhost and 1and1, my router configs, serial console servers, and some other backups. I was also replicating some directories on zen to orac as backup (left over from when zen was a Windows 7 PC....which saved me from losing everything when it scrambled itself.)
But, back to my printer quest.
So, the announcement of FreeBSD 9.2 came out on Monday [September 30th], which I missed because I was focused on my UNMC thing. But, once it appeared, I knew that I was going to want to upgrade to it sooner than later.
From its highlights, the main items that caught my attention were:
But, I did start this upgrade on October 4th....where for an unknown reason, I launched the
freebsd-update process on cbox, the busier of the two headless servers. I suspect I went with doing the upgrade on my headless servers, because they are entirely running on SSD and would likely see the benefit of lz4 compression. And, perhaps I did cbox, because it was the system that could most gain from lz4.
It took a couple iterations through
freebsd-update, before I got an upgrade scenario that could proceed. And, it took a long time given the high load that is cbox.
That is cbox is an Atom D2700 (2.13GHz, dual core) processor. And, cacti (especially with the inefficient, processor/memory intensive percona monitoring scripts -- might help if only scrpt server support worked, and wasn't just a left over from what it was based on.) being the main source of load. That is usually in the 11.xx area, except during certain other events (like, since 3.5, when
cf-agent fires...cbox is set to run at a lower frequency than my other systems.) or when the majority of logs get rotated and bzip'd. And, there's also some impact when zen connects to
rsyncd each day for
backuppc. But, these spikes weren't that significant. Though the high load would cause
cf-agent runs to take orders of magnitude longer than other systems, including its 'twin' dbox.
Also ran into a problem (again?) where a lot of the differences that
freebsd-update needed resolved were differences in revision tags....some as silly as '9.2' vs '9.1', others had new time stamps or usernames, but seldom any changes to the contents of the file. Which I then discovered a problem from having some of these files under
cfengine would revert these files back to having '9.1' revision strings, which confused the
freebsd-update. I ended up updating all the files in
cfengine to have the 9.2 versioning, though I thought about just removing/replacing it with something else entirely, though wasn't sure the impact that would have on current/future
Though it did seem to cause problem with the other two upgrades, where it would say that some of these files were now removed and asked if I wanted to remove these. Which doesn't make sense, since it didn't say that with the first upgrade. It was probably just angry that these files already claimed to be from FreeBSD 9.2.
It also didn't like that I use
sendmail, therefore my sendmail configs are specific to my configuration, or that I use
printercap is the one auto-generated by cups, etc.
But, once it got to where it would let me run my first "
freebsd-update install". I ran it, rebooted, ran it again, rebooted, updated stuff (though it didn't complain as much, perhaps because some of the troublesome kernel mod ports had corrected the problem of installing into
/boot/kernel, or perhaps enough stayed the same between 9.1 and 9.2, that things didn't freak out like before. And, this includes the virtualbox kernel mod, when I did the upgrade on zen, and later mew. But, I re-installed these ports and lsof. I did a quick check of other services, and then upgraded the 'zroot' zpool to have feature flags (which now means it no longer has a version, apparently instead of jumping the numbers to distinguish from Sun/Oracle it has eliminated having version numbers (for beyond 28) and having flags for the features added since. Wonder if the flags capture all has changed since 28, since I thought there have been other improvements internal that aren't described by version numbers. Namely, I seem to recall that there have been improvements in recoverability....namely it had been suggested, when I was trying to recover a corrupt 'zroot' on
mew, to try finding a v5000 ZFS live CD. Which I don't think I ever found, and gave up anyways when I concluded the level of corruption was too great for any hope of recovery and that I needed to resort to a netbackup restore, before the last successful full get's expired. Though being that it was nearly 90 days old, the other two month fulls didn't exist due to system instability that eventually caused the corrupted zpool (eventually found to be a known bad revision of the Cougar Point chipset and a bad DIMM...things seem to finally be stable from using a SiI3132 SATA controller instead of the on board, and getting that bad DIMM replaced....was weird that it was a Dell Optiplex 990, purchased new over a year after the problem had been identified and a newer revision of the chipset was released. I did eventually convince Dell support to send me a new motherboard and replace the DIMM. The latter was good, since I had to use DIMMs from another Dell that had been upgraded, so I had less memory for a while. But, while at first I did use the onboard SATA again, eventually I started having problems that would result in losing a disk from the mirrored zpool, to eventually causing a reboot where they would both be present again [though gmirror would need manual intervention]....and moving back to the SiI3132 has finally gotten things stable again. Though the harddrives in mew are SATA-III, so it would've been desirable to have stayed on the SATA-III onboard ports, where it was these ports that were the main source of problems in the prior defective version. Perhaps the fact that the prior version had a heatsink and the new version didn't, wasn't because they didn't need it to try to compensate for the problems caused by over-driving the silicon for the SATA-III portion. But, an oversight with the newer revision motherboard. The problem did tend to occur in the early morning hours on the weekend, when not only is there a lot of daily disk activity, but there is also a lot of weekly disk activity, etc. Oh well.)
So, after upgrading the zpool, and reinstalling the boot block/code. I then rebooted the system again. I had already identified the zfs filesystems where I had 'compression=on', so had written a script to change all these to 'compression=lz4'. Which I now ran.
And, then I turned my attention to doing dbox.
Pages: 1· 2
So, there's this BOINC project out of Poland called Radioactive@Home, where you have a radiation detector hooked up to a computer taking samples, etc. Its my second BOINC project with a hardware sensor. Though I had signed up for this one first...back on June 16, 2011. QuakeCatcherNetwork had come later, but getting a sensor was quick (though there were delays in getting it working, they had switched to a new sensor where they didn't have Linux drivers yet...etc., etc.) But, doing Radioactive@Home took longer as sensors are built in batches, there had been early batches that I missed and I wasn't all that sure at first if I really wanted do go to the hassle of getting one.
But, then another user announced that he would do a group purchase of 50 or so, which it should cut shipping costs quite a bit by having a cheaper large shipment from Poland, plus domestic delivery for the last leg. The way delivery costs go, you can get up to 3 for the delivery charge...though most people only want one....at least initially.
Basically I ordered my first detector around August 2011, and finally received it in March 2012. And, it just runs...though occasionally I'll look to see if anything interesting is recorded (like the interesting trace for around the end of the world....)
Meanwhile, on June 26, 2012 there was an announcement of a new detector...a pretty looking one. My first sensor was a prototype type case with rough cutouts, etc. Not really bad looking, but still plain and crude looking. While the announced sensor looked neat, the kind of thing that I might considering putting on my desk at work....
So, there was basically an announcement that there wasn't going to be another bulk US purchase...so after some thought, I decided this new detector was just too pretty to pass up. So, I ordered one mid to late July, 2012. Got confirmation on July 23rd, 2012. 27 Euros for the detector plus 10 Euros for up to 3 detectors, more than 3 pay for the detector now, get bill for actually shipping cost later. Plus if I use PayPal to specify that I'll pay the transaction fees....
In the previous order, it had been requested that we have PayPal funds to pay for the transaction....or use a check. I had tried to keep a float of cash in my PayPal account....but when it finally came time to pay, there wasn't quite enough to do that, so I opted to just mail a check. For this second order, I went with PayPal and had PayPal add the transaction charges to my total.
First detector cost me $46.25 by personal check. Second detector cost me $47.36 (and conversion and including the transfer charge).... I sent the PayPal money on August 21, 2012.
And, then it was wait and wait and wait. I would check the boards now and then for updates...but it was mostly other people wondering the same thing.
Eventually, I stopped checking in...and kind of forgot all about the sensor. Though I did visit the site briefly, but didn't linger or read the detector threads...which I went to check what platforms the project supported. Because when I had originally ordered, I was down to a Solaris 10/x64 workstation, a Windows box, a first gen MacBookPro (32-bit Core Duo). and a dead Linux machine. Eventually, I got a computer to replace the dead Linux box...but I went with FreeBSD instead, and it eventually displaced the Solaris workstation. In February, 2013 while I was working late on my FreeBSD system, I saw the Windows box update itself and reboot, and then it failed to boot. It had killed itself....pretty much the same way my home Windows box had killed itself in an auto-update in February, 2012. I left it off, not sure what I would do with it....I thought about OmniOS or SmartOS...though it was a first gen i7, so no EPT for KVM. Eventually, I decided to install Ubuntu 12.04LTS on it....where its mainly backup for when my FreeBSD system crashes.... its one thing that new Seagate drives only have 1 year warranties...its another thing that they seem to have trouble lasting that long.....
And, then an iMac 27" appeared on my desk....back when it seemed bleak on getting FreeBSD working as my main workstation....I was talked into getting one. But, FreeBSD remains my main workstation....while there are somethings that the iMac is the only computer I have where things work (like being able to participate in WebEx, Lync, Google Hangouts or Xoom for web conferencing....plus it finally solves having mail staying open while I switch to the appropriate desktop to do whatever....I'm up to 17 now....where there are typically 4 to 12 windows...either of uniform size, or variable size, and some desktops the windows overlap, though that desktop is mainly for tailing logs.... Where I'm up to 2 full desktops and 2 half desktops for that.... Anyways, I had made a quick visit...because I wondered if Mac OS X was a supported platform (it wasn't) or if anybody was using FreeBSD for this project....didn't get any search hits. And, it seemed unlikely that the hardware part would work through the Linux emulation on FreeBSD (especially the Fedora 10, and I'm not sure what the process for converting to the CentOS 6 is, that wouldn't break all the things I'm using Linux emulation for....though it is mostly other BOINC projects.) Though doing the search now, I see that a couple days ago the question got raised....with not much luck on having it find the detector ... but ending with a link to a FreeBSD version of the application.... Though since I have a Linux system at my desk (where is primary purpose is to run VBoxHeadless containing Windows 7, for those occasions where I need to use vSphere Center...and passing the time doing BOINC)...I'll just go with running new detector should it ever arrive...on that.
And, its the same time suck....cacti.
Last weekend got away from me, because I to make another attempt to improve
cacti performance. I had tried adding 3 more devices to it, and that sent it over the limit.
I tried the
boost plugin....but it didn't help, and only made things more complicated and failure prone. Evidently, updating rrd files is not a constraint on my
cacti server. Probably because of running on an SSD.
I made another stab at getting the percona monitoring scripts to actually work under script server, but that failed. I suspect the scripts aren't reentrant, because of their use of global variables and relying on 'exit' to cleanup things it allocates or opens.
I had blown some previous weekend when I had tried to build the most recent version of hiphop to maybe compile the scripts, but after all the work in figuring out how to compile the latest 2.0x version...it would SEGV, just as the older lang/hiphop-php did after resolving the problem of building with the current boost (a template had changed to need a static method, meaning old code won't link with newer boost libraries without a definition of this.) And, this is beyond what I have in my wheelhouse to try to fix.
During the week, I had come across some more articles on tuning FreeBSD, namely a discussion of kern.hz for desktop vs servers. Where it being 1000 by default is good for desktops, but the historical setting of 100 being what to use for servers. Though IIRC, ubuntu uses 250 HZ for desktops and 100 HZ for servers, it also doesn't do preemption in its server kernel along with other changes (wonder if some of those would apply to FreeBSD?) Though modern kernels have been moving to be tickless. Which I thought was in for FreeBSD 9, though the more correct term is dynamic tick mode...and which is more about not doing unnecessary work when things are idle. Which isn't the case with 'cbox'. So, perhaps, fiddling with kern.hz and other sysctls might still be relevant. Though haven't really found anything detailed/complete on what would apply to my situation.
So, I thought I would give kern.hz=100 a shot.
At first it seemed to make a difference....no improvement in how long to complete a poll, but the load was lower. Until I realized that a service had failed to start after reboot. I had only run the rc script by hand, I hadn't tested it in a reboot situation. And, its not an rc script....it was used to be a single line in rc.local that worked on ubuntu and FreeBSD (except on one of the Ubuntu systems it results in a ton of zombie processes, so making it an init.d script that I could call restart on happened.
So, I spent quite a lot of time reworking it into what will hopefully be an accept rc script. One thing I had changed was that instead of using a pipe ('|') which was causing the process after the pipe to respawn and turn the previous process into a zombie each time the log file was rotated and "tail -F" announced the switch. And, this was while I was moving the service to FreeBSD (and management under cfengine 3.)
Though looking at my cacti graphs later....while the service had failed to start after reboot, it turned out to have been running for sometime, until I had broken it completely in trying to rc-ify the init script. Will, duh....I had cfengine set to promise that the process was running, and it had repaired that it hadn't started after the reboot.
Another thing I had done with I had init-ified the startup of this service, was I switched from using pipe ('|') to using a fifo, which addressed the respawning and zombie problem and eliminated the original reason to have an init.d script....
While the init.d script had worked on FreeBSD...it was just starting the two processes with '&' on the end then exiting. FreeBSD's rc subroutines do a bit more than that. So things weren't working. The problem was that even though I was using daemon instead of '&', so that daemon would capture the pid and make a pidfile. seems daemon wants the process it manages to be fully working before it'll detach. But, the process is blocked until there's a sink on the other end of the fifo. (does sink fit was the name for the fifo's reader?) I first wonder if I could just flip the two around, but I suspect starting the read process first would be just as blocked until the write process is started. So, I cheated by doing a prestart of the writing process and only tracking the reading process.
Though it took a bit more work to get the 'status' action to work....eventually found I needed to define 'interpreter' since the reading process is a perl script. And, the check_pidfile does more than just check to see if there's a process at the pid, but that its the right process. And, it distinguishes between arg0 and the rest.
Pretty slick...guess I need to do a more thorough reading of the various FreeBSD handbooks, etc. Of course, it has been 13+ years between when I first played with FreeBSD to its take over of my life now.
As for the tuning....it had made a small difference, but no improvement on cacti system stats. Basically the load average fluctuates a bit more and the CPU utilization seems to be a bit lower...though it could because the 4 lines of the cacti graph aren't so close to each other now.
Meanwhile...I noticed that one of the block rules in my firewall had a much higher count than I would expect, so I think I was about to get logging configured to see what that's about.....(which I was working on when I remembered that I hadn't rebooted after making the kern.hz change to /boot/loader.conf yesterday...the commit also picked up files that I had touched while working on moving the one remaining application on 'box', though that may get delayed to another weekend....perhaps the 4 day one coming up.)
I had set cf-execd's schedule to be really infrequent (3 times an hour), because I was doing a lot of testing and cf-agent collisions are messy....messier than they were in cfengine 2 (in 2 it usually just failed to connect and aborted, in 3 it would keep trying and splatter bits and pieces everywhere....which is bad when there are parts using single copy nirvana. resulting in services getting less specific configs, until the next run.
But, I sort of brought back dynamic bundle sequences.... but key off of "from_cfexecd", so I can test my new promise with less problems of colliding with established promises. Though there are other areas where things still get messy.... need to clean up some of the promises I had based on how things were done at work, so that the promises are more standalone.
Kind of weird using my home cfengine 3 setup, and other admin activities, as the means to break the bad habits I had picked up at work....
I had acquired my first HSTI Wireless Media Stick back on April 24th, 2011 (from a marketplace seller on Amazon.com)...it took some time to arrive and I blogged about it on April 30th, 2011 -- Getting local content to show on my Roku XDS.
Now my Roku has moved to my other HD display (24" 1080p), but that was before my old Samsung HDTV (43" 720p) regenerated into a Samsung Smart 3DTV (50" 1080p).... so I'm back to living room TV being my main viewing device for all content, though TiVo has a box that I could connect to the smaller display to access my TiVo content there....which I may want to get at a later date. The majority of content I watch is from TiVo...one of these days I need to setup my blu-ray player so I can get back into watching DVDs (not sure when I'll have blu-ray discs, but need to get back on my netflix backlog).
But, the other day I had an mp4 file that I needed to play....and I thought I should get someway to do that to my 50" HD display... Had to settle with using the Roku for a bit. And, decided that the plan will be to acquire another HSTi Wireless Media Stick.
After searching around online, eventually found that ordering directly from HSTi was the only option now. So, I ordered another one on May 17th. It arrived yesterday. But, I didn't set it up until I got home from work today. Somehow I had forgotten again that HSTi is in Calgary, Alberta. Not that I'll be going up there in the immediate future....
Anyways, no big surprises...good thing I had solved my USB2.0 and Windows 7 in VirtualBox on FreeBSD problem (got a Silex SX-DS-4000U2). I'm sharing TARDIS from orac to it still, since I don't yet have a replica on zen yet (need to free up space). Though when I moved the HSTi Wireless Media Stick it had forgotten the share, so had to pull up web again and add it back. Interesting that its graphic shows the itself, while the graphic on my older stick is that of the original Wireless Media Stick (it used to be the correct graphic, but after an update it keeps showing the graphic of the older version.) Though this one came from the factory with the latest firmware, so who knows what'll happen when there's an update.
Was interesting using the SmartTV to view it, though wonder if it'll be a problem with it constantly discovering the stick every time I turn it on and presenting dialogs and such. Afterwards I tried the Amazon app to see if that was working yet....it was still saying I needed to update my TV, though this time there was an update....and now that works. Which might make it interesting to decide on what I should do. The only problem with using the SmartTV versus some other viewer....the TV is only 2.1 audio while other routes I can get 5.1, and its a different input on my receiver....
Oh well, back to other projects....
So, doing cacti on cbox doesn't seem to be working long term... but, the moment is being prepared for....I starting to assemble the pieces to build a new machine to do this and handle some other tasks that I've been looking for a place for.
Back to cfengine, I added a promise for dnetc (distributed.net)....and then a promise to finally configure CUPS on the two servers. And, then I turned to nagios.
I spent a couple evenings creating the initial configuration of nagios, working in design changes that I wanted to make and initial monitoring of localhost (dbox). Though it wasn't straight forward....there were differences here and there....mostly in FreeBSD layout, paths, and some of the commands taking different options. But, eventually I got everything running. My old check_dyndns worked once, but then stopped working.... problem was that it did 'stat -c "%Y" ..." which doesn't work on FreeBSD, 'stat -f "%m" ...' was the adjustment for that. All, while all the checks_* seem to be there, command definitions was lacking....but I guess having command definitions for everything is part of the debian/ubuntu packaging. There were other frills that came with that, that I don't mind not having...
I did run into check_ntp being deprecated....with check_ntp_time and check_ntp_peer being the tests to use....separating and making more clear on whether you're comparing time between servers using ntp or checking the state of the ntp server...
It did show some interesting oddities in holding NTP time on my home network.... I know that I should have 3 or more ntp servers, but it seems that I'm often landing in the state where I only have 2....with lots of delay, resulting in pretty good swings of jitter....almost makes me wonder if this something I could graph in cacti....
Wonder if I can find a cheap NTP appliance somewhere....
The last stumbling block was check_dhcp. Which seems to be broken on FreeBSD. All, the discussion on it seemed to point to firewalls, but no firewalls and it still didn't work....tcpdump on both places, and its saying it sending stuff, but no packets appearing on the network. But, I can see the other DHCP traffic on the network.
I remove that check and call it a night. I mull some possible work arounds....first one I tried was setting up linux compability and try running the check_dhcp from my working (ubuntu) nagios. Well, it didn't work...it couldn't find an interface. Oh well, guess there's the ugly way....use nrpe to invoke it. Though that didn't work right away.....probably because while I had created new nrpe configs for all my servers in cfengine, I haven't put any of my ubuntu servers under cfengine yet. Most of the other promises haven't been implemented for ubuntu yet. It was pretty simple to include nrpe.cfg for everything.... in fact it condensed to only 3 files.... a freebsd version, an ubuntu version and a host specific version for orac. Well, not right away...that happened more recently...while I was going through and updating the nrpe.cfg's by hand on the ubuntu servers. Was when I noticed that some of the files were only different in comments....so I made further simplifications in cfengine...which'll propagate out eventually....
Long term, I'll probably just have to track down some alternate implementation of check_dhcp....
I then add cbox to monitoring...and then looked to see about monitoring things that are on cbox/dbox...so I found checks for freeradius, cups, squid, along with improvements to checks on ntp. The check_squid was tricky....I got it working by hand, after making the suggested change for the default Cache type parsing, which turned out to be changes for squid3 vs. squid2 (but box is still running squid 2.7 - since I had re-built it by hand with SSL support, and blocked ubuntu from updating it. Orac wasn't blocked so it eventually turned into squid3.
it worked by hand, but wouldn't work under nagios...turned out that the embedded perl wasn't liking it. I was going to disable embedded perl for it, when I took a look at seeing what it was complaining about. And, did some reading on embedded perl.... the gist was "use strict", "perl -w" and "perl -c" as starting points. perl -w was find, but perl -c had one problem....which I fixed. But, no go. And, then noticed the line "# todo : use strict", guess I'll have to deal with that.
And, making that all happy, got it working.
The only other quirk was the memory check wouldn't work on FreeBSD, I guess there's no mallinfo() available for that. So, no running that test on those servers....plus no Cache test on box. But, it still left enough variety of tests that worked on all. And, it wasn't so much that I wanted to get all the information, but I choose to define all the different tests with ports set into the test....so running the check would also test that all my squid ports worked. There's actually only two that matter, but I have all my squid's configured the same, listening on 5 or 7 ports....depending on whether I have SSL enabled. Though I pretty much only need two now. I'm not doing transparent proxying and I don't need the SSL now that I've split box into dbox/cbox....the SSL was so ddclient could work on box and update dyndns via proxy to DSL....
Next up is adding zen to nagios, and coming with with more tests of things that are specific to zen, but covered or not covered in the old nagios.
Though as I worked along...there were things I couldn't find monitors for...though I realized that I could have cfengine promise that those services were running. Plus cfengine was also taking care of other things. So, I should probably work on writing some promises for zen. So, I can have promises to make sure things are started up again after a port is updated or that php/extensions.ini is reordered, etc.
But, I'll probably continue adding everything else to nagios first.
My 10.04LTS servers are nearing the end of the line, but the move to get them upgraded is starting to get close....
At first I was waiting for the 12.04.1LTS update to appear, so that I might try upgrading them to Precise. But, then it struck me that it struck me that it might be time to replace one of the servers with new hardware.
The boxes had been servers for many years now, but in 10.04 they dropped support for 32-bit hardware as servers....which means one of my servers, and the more important one...is nearing EOL.
So, back in August....I came across a Shuttle XS36V open box on newegg.com....so I ordered one, along with an 8GB SODIMM memory set (a pair of 4GB)....the computers support a maximum of 4GB, but in my mind I was thinking that if this worked, I might get another to make a pair.
But, then it sat around as I go on to other projects....I had figured on using one of the 120GB SSDs that I had acquired earlier....
Realized what was strange about these boxes was they had serial ports, so I started thinking about network serial and setting these boxes up headless. Since I had done a server at work for serial with FreeBSD, I got to thinking that I would do the same with this. Plus I was really finding to be much more predictable/stable than Ubuntu....something that's good in a server.
I started revising the plans as I went....ended up deciding that I would have one that is default cable and one that is default dsl, and running the usual core...one would be master DNS, and the other one be the primary slave....just as one would be primary DHCP and the other secondary....both would do freeradius. Both would NTP servers. Both would do mail, though zen is my main mail system. They would have both apache and nginx on them...apache to run local web services....and nginx doing reverse proxy to the other web apps on my home network (kind of important since current cable router is a step back with it only supporting 16 port forwards....though I have a replacement that I'll get to setting up some day...)
Then while I was poking around on newegg.com, spotted that there was another open box Shuttle XS36V....so now I had two boxes. I had picked up once Cyclades ACS1 earlier on ebay, so I got a second....a 'new' one for about the same price. Just needed to find time. Perhaps after FreeBSD 9.1 drops.
Well, it dropped....but I still wasn't ready...now I needed some SSD drives for the machines. So, I used the $30 in RewardZone for the new TV I had gotten in September, and picked up a pair for 128GB SanDisk Extremes. And, then I started setting up the ACS1s....hmmm, wonder if I have any serial cables. Okay, I'll buy some of those first.... I end up ordering from TigerDirect, as they have a better price on RJ45 to DB9 adapters....not that I need them for the actual ACS1 to Shuttle XS36v connection...that's just a 9F to 9F null modem cable.
I have a number of FreeBSD projects at home on the go that I've been trying to find enough free time each weekend to tackle. Having a recent problem where my system will freeze/panic/reboot whenever I shutdown or saved a VM...I thought maybe upgrading to 9.1 would help. It didn't.
Anyways...I wondered how freebsd-update would proceed, given that checksums on everything had changed...because at one point during things I had rebuilt the world with optimized compile flags, though otherwise still running GENERIC and no other alterations, though later rebuilding kernels so uname would reflect p4 and p5....
But, from what it said it was going to do to get me upgraded, it seemed okay with everything. Though the merge was a bit strange....first time through it didn't present one of the merge conflicts for me to resolve (/etc/login.conf), so I said no when it asked if the file was ok before proceeding. That just stopped it instantly, had hoped that since it didn't ask how to resolve before that it would ask now. But, running the update again....it did ask for that file this time (otherwise I was considering what would happen if I fixed it by hand during or later.... So, then it went and upgraded.
And, then that's where all the fun started. While rebuilding all my ports wasn't necessary. It was necessary to rebuild all the ports that create kernel modules....like nvidia, virtualbox, fuse, webcam....
Over the last year, I started buying SSD drives. It used to be that they seemed pretty expensive, and of questionable performance and reliability. But, all things have improved over the years. And, when 120GB drives dropped under $1/GB (initially after rebates, later before rebate [if any]). I didn't have an immediate need for an SSD drive at the time, but I envisioned replacing the drive in my (u) laptop. And, perhaps my (w) laptop....beyond that I wasn't sure.
3/05/12 - Patriot Pyro 120GB - $159.99-$40 rebate = 119.99 3/20/12 - OCZ Agility 3 120GB - $139.99-$30 rebate = 109.99 4/27/12 - Mushkin Enhanced Chronos 120GB - $ 99.99 5/16/12 - OCZ Solid 3 60GB - $ 74.99-$20 rebate = 54.99 8/10/12 - Sandisk 128GB - $ 79.99 11/21/12 - Kingston HyperX 2K 240GB - $149.99 1/18/13 - Sandisk Extreme 120GB - $ 89.99-$15 reward = 74.99 1/18/13 - Sandisk Extreme 120GB - $ 89.99-$15 reward = 74.99
But, during this time...I had the lhaven misstep Where I had picked up the 60GB drive for that, but ended up using the Mushkin 120GB drive instead. The OCZ Agility 3 120GB had gone in as OS drive for my Xen Cloud Experiment. And, stayed when I went on to making it FreeBSD. Cut up initially as 64k boot, 32GB swap, 16GB L2ARC for the mirrored 1.5TB drives to help with dedup...and the rest...63GB root zpool.
Somehow the Patriot got misplaced for a while, so it got overlooked during the chaos.
After a 'break'. I picked up the Sandisk 128GB drive. Thinking that it might be a better choice to use to finally replace the (u) laptop harddrive. Though I waited until after the NN conference in October to do it, but before I went to LISA in December. Though I didn't finally upgrade the the OS to 12.04 (from 10.04) until just last week. I had thought about doing a clean install to 64-bit....since there were some issues since I upgraded the memory to 8GB. But, changing the hibernate method seemed to have solved the issue....so decided to leave it 32-bit. Though my (w) laptop is 64-bit...though it only has 4GB. Not sure if I'll upgrade it to 8GB. Or when I'll upgrade its harddrive to SSD.
Things have been kind of tight since...on the 128GB Sandisk drive. partly because swap got a bit bigger...had suspected that 8GB swap was iffy for hibernation....so had bumped that up. Plus the original harddrive was 160GB. But, the lion share of space consumption is my Windows XP VM. But, it get's the job done.
Meanwhile...around this time I got the idea that instead of making the risky upgrade of my two Ubuntu servers from 10.04 to 12.04, that I would set up two new FreeBSD servers and migrate the essential services over before deciding the future of those systems. So, I acquired a pair of Shuttle XS36V's...4GB of memory for each...and then eventually the plan was to acquire a pair of SSDs for them. Which I finally did last week as a pair of SanDisk Extreme 120GB drives. (with the help of $30 from Best Buy Reward Zone....and this purchase should get me a $5 reward zone soon.) And, these will probably get installed as FreeBSD 9.1.
Pages: 1· 2
Latest Poopli Updaters -- http://lkc.me/poop
|<< <||> >>|
freebsd eyeglasses voip dvd backuppc ebay boinc ups orac cox prescription virtualbox tv «hd movie» lhaven dsl «instant streaming» «doctor who» «powersource 400» 10.04lts raid raid1 tardis batteries mdadm «windows 7» upgrade cfengine3 newegg amazon.com ubuntu zen usb b2evolution «air purifier» woot replaytv tivo «watch instantly» netflix «chicago tardis» «tivo hd» cpap box linux «amazon prime» «sans digital» appletv «windows xp» twitter