« first day (987 days earlier)      last day (3994 days later) » 

6:00 PM
what if you had some software installed on every pc and you had to remove it before you could install the latest version? what would you do right now?
 
Go to about 30 machines and do it the way I described previously
would be a suck day for sure
I've had them, like when windows live stopped working and I had to put skype on everybodies machine
 
again - write a script to uninstall the softwre and another to install the new version. and run it against every machine, sit back and relax
 
color me sold, that could save me a headache
 
there is puppet for windows too... just saying
 
Same type of thing?
I don't really use alot of automation tools at the moment
 
6:03 PM
you should
 
admittedly - the software would have to support unattended instllation, meaning you provide all the instructins either in a config file or as command line arguments. but i have found almost all software for windows systems supports unatended installation these days
all the stuiff i use anyway
 
I could prepose it
 
what?
 
The corporate network is the one where I actually have authority to make real changes
propose*
 
propose what?
 
6:06 PM
implementing some sort of solution for automated installations and such
although that might not be used much either
we keep their machines pretty minimal, just what they need to do their jobs
unless of course there was an update that required re installation
The only instance I've seen being the skype thing
 
well - i think you'd find powershell an interesting challenge. but im just trying to think of a few real life projects you could work on.
 
Yeah that's what I've been trying to do
The only excitement I get anymore is when we decide to add something to our network
Like the IDS thing, we never had that on the corporate network till I offered to set it up and monitor it
and WSUS for the secure network
 
ah ha!
here's an idea for you...
ms havvnt made it very easy - but i use powershell to install the updates on my servers.
 
I am assuming your company is also very opposed to auto updates ;)
 
you know how one set of updates install - a reboot is required and then sometimes another set of updates are needed. you can use poweshell to keep installing the updates even when a reboot is required until there r no more to install
 
6:11 PM
We can never do auto updates cause nothing can suffer unschedules downtime
 
i 100% agree
auto update should not be used on svrs
 
I see how that could be useful
 
thats why i use this solution.
 
Can you write custom messages to event log to indicate when the last round is finished
or anything of that nature
 
absolutely
or send an email
or write to database
or write to a logfile
 
6:13 PM
either would be useful
I keep a central syslog for corporate and secure
so one of my installation scripts that I just wrote lets me know if the server installation finished without errors by writing to syslog
Gone are the days of having to babysit an ubuntu server installation through every step
 
powershell has a very good restart command in version three which allows you to perfomrm a reboot, wait for the machine to come back up and then carry on with your script.
good move
 
sounds like script porn to me
Well I had to, I would always have to walk away during the long parts of the installation and guesstimate when to walk back into the datacenter to continue
seemed logical to have it alert me when all was finished so no time was wasted
 
how do you build your windows workstations?
 
We have an image
 
Role dependent, I know I will get body slammed in this room for not using an image, but the infrequency in which we need to install workstations has swayed me from that. I do it the old school way
Whoever their manager is going to be gives me a list of software they will need, assigns it to me in our internal work request system, and I do it
 
6:18 PM
I've never bothered much with imaging. Mostly because we have PCs of all different models. I use PXE (Pre-Execution Environment). The Windows setup is 100% automated as is the installation of all the software and lastly the windows updates.
I use Windows PE in conjunction with PXE
 
I've heard of PXE but never implemented it
I rarely have to install workstatios
ALTHOUGH
There is going to be a dark week in the near future where I will have to replace something in the neighborhood of 50 windows boxes in our datacenter with the cute little poweredge R210IIs
If I could make that easier I would
 
you're moving to VDI?
 
nar
 
oh sorry no - i misread
 
Just replacing old boxes with newer boxes
we were using towers for a particular set of servers back in the day, all installed with windows 7 and serving a single purpose, in a distributed fashion
 
6:22 PM
why 50 physical servers? do you use virtualisation?
 
now we are gaining back real estate by using the 1U poweredges
no virtualization
 
oh. why not?
 
My boss probably has his own reason, but I am assuming that he does it for fault tolerance or something
If one physical server dies, it's not hosting anything but whatever application it is desinged to serve
And he explained something to be before that I was a little to inept to absorb at the time, but I remember him saying something about horizontal scaling vs virtical scalig
So as we expand, we can simply add more devices to handle the distrubuted workload, rather than continue to upgrade one or two mega servers to handle the increase
I was new at the time but it made sense to me
 
Not flawed but old fashioned
 
6:27 PM
Unless all those physical servers run at full load all the time, that's such a waste
 
He's been doing it that way for like 12 years lol
They pretty much do
 
Yeah, old school.
 
well - keep one of the old servers, install hyper-v or vmware on it and setup a virtual test environment.
 
What does that do in the way of fault tolerance though
 
my previous suggestion wasnt aimed at improving fault tolerance - im just suggesting a little project to get your hands dirty with virtualisation
 
6:30 PM
Well, when a single piece of hardware fails (which you'll have fewer pieces of hardware, so fewer parts to potentially fail), your virtual machines with application workloads will be automatically rebooted on a working piece of hardware instead of staying down.
 
It would be fun
 
And if you need to do maintenance on one piece of hardware, you can move the virtual machines off to different hardware without taking them down.
 
And you could do this seamlessly with minimal downtime?
 
@shane - do you use vmware or hyper-v?
 
@Miguel No downtime. Milliseconds of network connectivity loss for the switches to figure out where the MAC address moved to.
@Fitzroy VMware mostly.
 
6:32 PM
i mostly use hyper-v
 
See that is what sucks about my job
I have little to no knowlege of some of the newer technologies
and methods
 
we have a few dev environments that use hyper-v server 2012
 
Still though I am unsure of how it would be implemented in my environment
 
not win svr 2012 with hyper-v role - but the 'hyper-v svr 2012'.
 
@Miguel Yeah, understandable. If you can't get exposed to the stuff at work, mess with it in your spare time :)
 
6:34 PM
thats why i suggsted powershell as something to learn. you dont need to setup tons of VMs or need any physical kit.
 
I have like 10 servers at my disposal that are doing nothing
 
@Fitzroy Ah yeah, that's a good one.
 
I intended to use them for failover testing, sort of a live disaster recovery drill
 
@Miguel Nice! If you can, get the ok from management to set up a lab.
 
already did, just haven't had time
 
6:37 PM
are the old servers still going to have windows licenses or are you transfering those to the new servers?
 
I am interested in virtualization though, since it seems to be the thing to do nowadays
 
if not - you're gonna have to find a nice linux project!
 
they are bare, I smashed the hard drives already
 
oh no!
why?
 
Well, you can also fire up trial versions of Windows for short term lab projects like if you wanted to set up Hyper-V.
 
6:38 PM
Security policy, since they once had sensitive data, upon retirement the hard drives are destroyed
Although I have tons of spare unused drives to play with
 
i see
 
so that isn't an issue, but licensing is
 
as shane says you can use trial versions of windows
the server versions seem to last ages, 100+ days at least
 
We have reached a point where we are trying to overhaul and modernize our setup that has remained for the most part unchanged for quite some time
but it is a slow process, and done one piece at a time
we didn't get rid of our last PIX till last year
120 days for server I believe
 
also, if you use 'Hyper-V Server 2012' its free! BUT you hve to license every guest. BUT..if you run all linux guests you dont have to pay a penny.
 
6:41 PM
OUr environment doesn't allow it, or at least not all of it
Just to paint a picture, the majority of our servers are there for a single purpose, as part of a distributed workload
 
im not suggesting any prtactical usages here - just things you can mess about with
 
yeah but me being the curious type I am now trying to work it out in my head lol
before I ever do anything I always spend assloads of time brainstorming, trying to piece the puzzle together before I pull it out of the box
 
that approach will serve you well
always plan first
 
Yeah now that I think about it I doubt it would be feasable
for example
 
not once you've fowled something up
 
6:45 PM
The windows boxes I mentioned all exists soley to run a program, this program runs an instance of itself on each CPU core, through some programatic witchcraft
the program uses each core seperately to process incoming files, that are randomly selected from a central server, thus distributed
So with 50 some odd windows boxes with 4 cores per CPU, to try and stick all of those on one server and actually have the resources ( 50 X 4 cores each) would be very difficult at best
I am assuming that VM's let you allocate cores and other resources on install time
Or am I incorrect about that
 
@Miguel No need for you to run it all one one server; you'd essentially have a cluster of physical hosts running your virtual machines distributed between them.
 
1x vCPU does not have to equate to a core
 
not core per say but threads
 
you can have more vCPUs than cores
 
Forgive me if I am wrong but I am only vauguely familiar
So if you have 2 physical cores and 2 vCores, the program can run 4 instances
the question being, are VMs capable of allocating something so specific for each VM
 
6:50 PM
sorry - thats not quite what im saying
 
oic
 
@Miguel You'll assign a certain number of virtual cores per VM, and their execution will be scheduled on physical cores by the hypervisor.
 
I told you I am not pretending to completely understand what I am getting at, it's just the best way I can explan it :)
@ShaneMadden In that case couldn't that potentially cripple the processing power of the individual hosts?
 
say you have two physical cores. You could have a 1:1 ratio or cores to vcpus but you do not have to
you could have two cores but four vms with one vcpu each
 
Currently the CPU on each physical host (50+) are maxed out during processing
 
6:52 PM
@Miguel If you overcommit (assign more virtual cores than there exist physical cores to run them on), then there can be contention between the virtual machines for CPU time, yes.
 
well more is better in this case, one host with one core is not as good as one with 4
It sounds wierd but this is one of those scenarios where every environment differs
 
that may not be for the right reason though
eg if your program is only single threaded - you have no choice but to run it once per cpu/core.
if its multithreaded you may find that one cpu/core would do]
 
It is single threaded, so it is running one instance per cpu/core
That much we know, and my boss is currently coding the new multithreaded release
 
@Miguel Right - and a modern virtual machine host might have something like 4 CPU sockets, each with a CPU that has 10 cores with hyperthreading - lots and lots of execution threads available.
 
I told you changes areslow lol
That would be so cool if I could talk my boss into it, he listens if I know exaclty what I'm talking about when I pitch an idea
 
6:56 PM
my point being - that the utilisation of each core by your program might be miniscule. So once you make it multithreaded you might actually get away with one core...
 
But the way your explaining it sounds seems feasable
Ok so if your up to it on to the next
 
Yeah, it depends a lot on the workload. If they ever back off their CPU usage (like when waiting for a new file?), then you might be able to allocate more vCPUs than you have physical cores in the system.
 
I have I think 60 or so linux boxes that run mysql and are a distributed workforce for searching
their usage is all RAM, and each of them have 24GB installed and use most of it
hard to see how I could shrink that down without spending more money than we already are
not to mention each of the servers themselves are 1U so the real estate isn't an issue there
Is there some situations where virtualization just doesn't make sense, and the old school horizontal scaling method actually makes logical sense
@ShaneMadden yes when they are waiting for files the cpu usage decreases dramatically
 
I would say there probably are cases - but unfortunately those are never discussed .
 
@Miguel Search? What kind of data? MySQL isn't typically the best for a straight-up search workload.
 
7:00 PM
In a nutshell
it's an OCR type thing
 
@Miguel Yeah - pretty much when you're Facebook or Google.
 
unfortunately - in IT, as soon as something new comes out, everyone turns nasty on the old way of doing things. Sometimes we have the sense to go back though.
take VDI for example
 
I can explain that
 
a few years ago thin clients were the future
 
@Miguel Maybe have a look at some of the distributed search stuff out there, like elastic search.
 
7:11 PM
The searching is for OCR. So the documents come in and are initially stored on one of a series of NAS servers. A copy of the document is sent to one of the windows boxes I talked about to get the text of the documents parsed and associated with the file by an ID. Once this is done the results get sent to a large linux db server which has the information on every document, the text associated with each document ( for text searching) and the location of the actual document on the NAS servers.
When somebody searches the system for text within a document, rather than massive amounts of queries htting that one linux servers, we hand a section of the results to one of the other linux servers. So now when somebody looks for text on a document, the results can come from one of 60 servers, which each server gets enough queries a day to make 24 GBs of RAM just barely enough to return the results within subseconds.
never heard of elastic search
 
7:23 PM
elastic search is fun
but if you use it: make sure you understand the JVM.
 
Not in my expertise toolkit as of yet
I figured out the correct term for what I was describing though
A database shard is a horizontal partition in a database or search engine. Each individual partition is referred to as a shard or database shard. Database architecture Horizontal partitioning is a database design principle whereby rows of a database table are held separately, rather than being split into columns (which is what normalization and vertical partitioning do, to differing extents). Each partition forms part of a shard, which may in turn be located on a separate database server or physical location. There are numerous advantages to this partitioning approach. Since the tab...
I'm not a database guy so I don't know it off the top of my head
 
haven't followed the discussion. Are you sharding your DB?
 
That's what it seems like
Although the term is new to me that is what I do
 
fun
say goodbye to unique keys and foreign keys :)
 
All I know is it is fast, but I couldn't see myself virtualizing it
 
7:26 PM
sharding and virtualization makes no sense as a combination
 
I figured as much
 
you shard when a single box is too small. Why the hell would you virtualize?
 
I can't think of a reason, I was just looking at different aspects of my architecture and seeing which portions of it would make sense to virtualize
Something I don't do at right now
 
what's your goal with virtualizing?
what problem are you trying to solve?
 
As of right now I don't know anything other than the fact that it seems to be all the rage. But with minimal research it looks as if it may be a good way to cut some costs and free up some real estate in the datacenter
 
7:29 PM
it's definitely all the rage, but that's not why you should do it :)
 
But it is not a need as of the moment
 
if you need to cut cost and reclaim real estate, it's a good thing to look at. If you don't, why bother with the hassle?
 
None in production, although I certainly would love to fill the knowledge gap with practical experience
 
that's a good goal :)
so you know what to do when the need arises
 
which I am sure it will at some point
 
7:31 PM
are you a windows shop, or something linux-y or BSD?
 
For the moment we can free up like two racks by getting rid of some gay towers and installing some 1U units that can do the same job better
Combination, windows for some tasks, ubuntu server for othres
 
then try both vmware and KVM. They both work in hybrid environments, with kvm needing a linux physical host and vmware having the snazzy tools
 
I've only used VMware for pentesting, never got nearly as involved as hosting servers
 
8:11 PM
KVM alone is quite spartan, people tend to use RHEV or OpenStack
 
kvm wilt virt-manager and virsh is ok
RHEV seems to be decent these days, now it no longer requires windows to manage linux machines :)
 
You all know my opinion.
RHEV was a disaster for us.
that was one year ago, though.
 
who's the 'us' there?
 
A trading firm I was working at.
 
8:21 PM
I basically left over the KVM versus VMware argument.
 
RHEV required Windows? Hah.
 
yes, for management
 
It has change a lot since that, though
 
yeah, even more than a year ago
 
If I were going to do a KVM deployment it'd definitely be openstack though.
 
8:23 PM
it has, but so has VMware. We looked at oVirt which was release shortly before...
 
We use our own scripts for deploying kvm. Still need to look at openstack and see if it would help us
 
no NFSv4 support in VMware yet?
 
@dawud No.
We get by on NFSv3.
 
@Dennis [this ](http://archipelproject.org/) project in still under heavy development, not for production, but take a look just for fun.
@ewwhite they promised support back in 3.x IIRC
 
@dawud That looks awesome.
 
8:26 PM
@dawud looks nice. Wonder how it integrates with our systems
 
yeah, looks nice. the demo is quite impressive, and hey claim support for KVM and VMware
 
I think I just like my virtualization solution to be a slam-dunk... straight-ahead. I don't have to worry about hypervisor stability on the VMware side. The kernel is refined well enough, supports the hardware I need... Good consideration has been given so I can PXEboot hosts or run off of SD card. Performance always improves with each revision.
 
@dawud that's just the start of what we need :) It needs to work with our asset mgmt system, ldap role based auth andnot want control of things like dns and monitoring
most 'server management' or 'vm management' software out there wants to control your entire environment, which is annoying to the point of not being usable.
 
Any look at onapp.com ?
 
@Dennis agree. our identity mgmt is an IPA server (includes DNSSEC and NTP), for monitoring, we use zabbix
Though I would really like to learn about AD and group policy in deep
 
8:32 PM
For instance, I want just one and exactly one graphing system, so I can easily correlate between any two metrics. So don't make your vm management thing do its own monitoring and graphing for crying out loud.
 
/me kinda a microsoft ignorant :(
@Dennis graphite makes that quite beatifully
 
@dawud indeed, I love graphite
we dedicate several dozen machines and many terabytes of disk to it :)
 
wow
@Dennis do you use [part of] the logstash~redis~mq~elasticsearch~statsd pipeline to feed graphite?
 
Being able to plot e.g. pageviews per minute and packet drops on memcached servers in the same graph is an insanely useful debugging aid (in this case it pointed out that the packet drops started well before the drop in pageviews, making it an unlikely cause of that drop)
@dawud we use diamond for symple measuring and the socket interface to pipe lots of custom metrics from our business monitoring system to it
we may be using stats as well, not sure
 
dinner time :) @all Gentlemen, always a pleasure.
 
9:20 PM
Roar
 
'sup lion king?
 
Lives. Circles. You know how it goes.
 
Hakuna matata.
And a bottle of scotch.
 
I could go for a nice bourbon drink.
I've never made a proper Old Fashioned. Perhaps I should seek one out.
 
WTF!
I need to get off of this site.
 
9:29 PM
@ewwhite what's the bike shop in Santa Monica called?
 
0
Q: ESXi 5.0 - Software RAID 1

LucretiusEnvironment: - Storage: HP P2000 MSA G3 SAS Array with 24 300GB 10k SAS Disks -- Two Storage Controllers with redundant SAS connections to each host - Hosts: Three HP DL380 G7s with a 10GB SD Card, CPU, Ram, etc... ESXi 5.0 is installed on the SD card in each host, this is the only local sto...

@JoelESalas Performance Bicycle.
 
Thank you!
 
@JoelESalas get some Wetzels for me
Desired result is ability to survive many disk failures. RAID6 gives me the ability to lose 2 disks per vDisk and still be okay. Applications that the servers run cannot go down. This is most important. Performance isn't much of an issue because the system is small. Right now I'm using Storage vMotion to move VMs off of a LUN with a bad disk, so that disk reconstruction performance doesn't affect performance. — Lucretius 24 mins ago
 
@ewwhite three and a half stars on yelp? Where are you sending me???
 
@JoelESalas it's a mega bike store... with a huge internet footprint
but they acquired the santa monica shop a few years back
 
9:34 PM
@ewwhite How can this compare to the spirit and moxie of a bike co-op in the hipster part of town
 
@JoelESalas It's clean... you won't need to use Purellâ„¢ after you leave...
 
@JoelESalas For one your bike will have more than one gear and made out of something lighter than 1/8th inch tubular steel.
 
oooooh!
Burn!
 
I aim to pleasure.
 
@ScottPack if YOU were a mod, how would you handle a contentious exchange like the one I linked to?
 
9:47 PM
If he's so scared of a triple failure that RAID6 isn't good enough, recommend RAIDZ-3? Heh.
 
@ShaneMadden if only... but the argument is stupid. I want to downvote.
:)
but I can't
 
@ewwhite Why not?
 
because that's a bitch move...
 
@ewwhite we still love you, even when you're a bitch
 
@DennisKaarsemaker Well, you have lots of HP disks die... any advice for the OP?
 
9:56 PM
didn't read the question. But fuck raid 5/6. 1+0 and spares is enough. I have lots of disks die because I have lots of disks. Rarely more than one in the same server and so far never lost a box due to too many disks failed.
did lose a box due to it being raid 0 and a disk got pulled because it said 'predicted failure'. That's just human stupidity though.
 
@DennisKaarsemaker In December, I had a triple disk failure on a proliant...
and had to rebuild very carefully... two failed disks and a predictive failure
the predictive died right as the array rebuilt
 
@ewwhite meh, I'd move that to a different box.
or just scrap the box if it's one of many
 
it was really a bad batch of drives from CDW
everything I bought at that time from that vendor failed around the same time
never forget.
 
not 'genuine hp drives'?
 
@DennisKaarsemaker Just a shit batch.
 
10:00 PM
shit batches happen.
We've had it with memory, which is more of a bitch to sort out than harddisks
 
10:12 PM
@ewwhite which road bike was your recommendation again?
 
@JoelESalas dunno.
go and see what they have.
I'm looking it up
 
afternoon gents
@dinn
 
@JoelESalas just search chat... or go
 
10:36 PM
@ewwhite that RAID61 guy works for Kroger. He really shouldn't have problems getting the budget for doing his service properly if it's all that critical to operations.
 
Oh fuck...
 
They only own about 30% of the grocery stores in the country.
There's 4 major chains in this area. They own 2 of them.
 
@Adrian My produce clients deal with Kroger... I've had to CALL Kroger's IT staff multiple times and walk them through Microsoft Exchange server settings because they would inadvertently do something that caused mail rejections.
 
@ewwhite Ah. And there's days that it really doesn't pay to live in Seattle. Too much talent here to compete with if guys like those can hold a job down. Too bad I'd never want to live in Atlanta.
 
They're not a sophisticated operation.
 
10:40 PM
@ewwhite They ought to be. They're the largest grocery retailer in the country and 4th largest in the world.
 
@Adrian Naw, it's like produce.
I've had to walk them through telnet email tests...
 
@ewwhite Weird. I've seen smaller organizations (Wegman's) that have more IT and business patents than they have stores.
Love me some Wegman's. Which I'm sure @jscott can sympathize with.
 
@Adrian Bad IT is everywhere!
2
 
@ewwhite Clearly.
 
11:47 PM
@ewwhite I went, they have some kind of promo going
 

« first day (987 days earlier)      last day (3994 days later) »