ML
    • Recent
    • Categories
    • Tags
    • Popular
    • Users
    • Groups
    • Register
    • Login
    1. Topics
    2. stacksofplates
    3. Best
    • Profile
    • Following 0
    • Followers 13
    • Topics 145
    • Posts 7,946
    • Best 2,861
    • Controversial 1
    • Groups 0

    Best posts made by stacksofplates

    • XenServer Import Disk Failure

      So I've been attempting to import a drive to XenServer for a while. Apparently you need the Windows 7 iSCSI initiator for it to work, which is blocked by GPO here. I found this, which I was able to use.

      From http://docs.vmd.citrix.com/XenServer/6.5.0/1.0/en_gb/guest.html

      The default transfer protocol is iSCSI. In which case, the Transfer VM requires an iSCSI Initiator on the XenServer host. An alternate transfer protocol is RawVDI.

      To use the RawVDI transfer protocol:

      Backup the XenCenterMain.exe.config file, which is located in the installation folder.

      Using a text editor, open the XenCenterMain.exe.config file.

      Add the following section group to the configSection:

      <sectionGroup name="applicationSettings" 
      type="System.Configuration.ApplicationSettingsGroup, System, Version=2.0.0.0, 
      Culture=neutral, PublicKeyToken=b77a5c561934e089" >
        <section name="XenOvfTransport.Properties.Settings" 
        type="System.Configuration.ClientSettingsSection, System, Version=2.0.0.0, 
        Culture=neutral, PublicKeyToken=b77a5c561934e089" requirePermission="false"/>
      </sectionGroup>
      

      To the end of the file, add the following section:

      <applicationSettings>
        <XenOvfTransport.Properties.Settings>
          <setting name="TransferType" serializeAs="String"> <value>UploadRawVDI</value> 
          </setting>
        </XenOvfTransport.Properties.Settings>
        </applicationSettings>
      

      Save the XenCenterMain.exe.config file.

      posted in IT Discussion xenserver iscsi vhd import
      stacksofplatesS
      stacksofplates
    • RE: What Are You Doing Right Now

      @Romo said in What Are You Doing Right Now:

      Trying to learn about the pros and cons of the various image formats (qcow2, raw, lvm) available for my kvm hypervisor. @stacksofplates any opinion and/or some benchmarks you can share?

      I don't usually use raw images unless I'm just testing something and don't care. Qcow2 is the slowest, but if you preallocate, it gets much faster. You can preallocate the metadata or do a full preallocation. If you do full though, I believe you lose thin provisioning.

      I don't have anything currently on volumes. I did for a while, but I don't have anything right now that needs that speed. I definitely go with volumes over raw files. Volumes give you all of the added benefits of LVM (snapshots, thin provisioning, etc).

      posted in Water Closet
      stacksofplatesS
      stacksofplates
    • Red Hat ReaR

      So for our hypervisors and our server for our tape library we want physical recoveries. Veeam has their beta endpoint restoration for Linux, but since we have RHEL for everything, I'm going to use ReaR.

      yum install rear genisoimage syslinux
      

      In /etc/rear/local.conf put:

      OUTPUT=ISO
      OUTPUT_URL=nfs://<share> or file:///<some-path>
      

      You can also use these options

      BACKUP=NETFS
      BACKUP_URL=nfs://<share>
      SSH_ROOT_PASSWORD="password"
      BACKUP_PROG_EXCLUDE=("${BACKUP_PROG_EXCLUDE[@]}" '/media' '/var/tmp' '/var/crash')
      NETFS_KEEP_OLD_BACKUP_COPY=y
      

      Then just run:

      rear -v mkrescue
      

      Then you can boot from that ISO to restore.

      Backups can also be taken with ReaR.

      In /etc/rear/local.conf add:

      BACKUP=NETFS
      BACKUP_URL=file:///srv/backup/
      NETFS_KEEP_OLD_BACKUP_COPY=y 
      BACKUP_TYPE=incremental 
      FULLBACKUPDAY="Day" ("Mon","Tues","Wed","Thurs","Fri","Sat","Sun")
      

      Last three lines are optional. By default it does a full backup, so if you want incremental then you need that line. And you will also need the keep old backup line for incremental.

      It can also keep the backup with the restore image. It won't be incremental though. For this you need:

      BACKUP_URL=iso:///backup/
      

      By default ReaR creates a tar image, however you can use rsync.

      BACKUP_PROG=rsync
      

      To create the backup run:

      rear mkbackuponly
      

      For backup and rescu ISO:

      rear mkbackup
      

      Make sure SELinux re-lables files on reboot

      touch /mnt/local/.autorelabel
      
      posted in IT Discussion red hat rear linux backup restore
      stacksofplatesS
      stacksofplates
    • RE: What Are You Doing Right Now

      Well I'm officially a RHCSA.

      posted in Water Closet
      stacksofplatesS
      stacksofplates
    • RE: Chrome Browser in Ubuntu 17.04

      @thwr said in Chrome Browser in Ubuntu 17.04:

      @thwr said in Chrome Browser in Ubuntu 17.04:

      @stacksofplates said in Chrome Browser in Ubuntu 17.04:

      While I usually agree, Chrome isn't in the repos, just Chromium. So if you need to install Chrome you can download the .deb and use gdebi, which will pull in all dependencies for you.

      You are right, Chromium != Chrome (they just share the same engine).

      But one shouldn't use a manually installed browser at all. They are a huge attack surface and should be kept updated, which is hard enough on Linux, but next to impossible for a Linux newcomer who installs a specific version outside of the package managers control.

      I think it's much better safer to either add Google's repo for installs and updates (https://www.google.com/linuxrepositories/) or just use Chromium. Either way, installing Chrome via a flat deb is usually a pretty bad idea.

      Not that I care about votes / likes / whatever, but -1 for this?

      Ya I don't know why either.

      Without a working internet connection, this is going to be very difficult unless you install from the ISO, but even then you might hit dependency issues with 3rd party packages. Chrome is probably the least of your concerns if you can't even get a working network connection.

      I'd go back to 16.04/16.10 and see if you can get the internet working on that first.

      posted in IT Discussion
      stacksofplatesS
      stacksofplates
    • RE: What Are You Doing Right Now

      Just got 12GB/s from a qperf test 🙂

      posted in Water Closet
      stacksofplatesS
      stacksofplates
    • RE: Chrome Browser in Ubuntu 17.04

      @travisdh1 said in Chrome Browser in Ubuntu 17.04:

      @Lakshmana said in Chrome Browser in Ubuntu 17.04:

      @scottalanmiller Full DVD means all iso in the ubuntu page?

      After a quick glance at the alternate downloads for Ubuntu, it doesn't appear that they have an "Everything DVD" like Debian provides.

      Ya you would probably have to make a local repo for everything. Never done it on Ubuntu, but it's relatively easy with RHEL.

      posted in IT Discussion
      stacksofplatesS
      stacksofplates
    • RE: What Are You Doing Right Now

      So my dad got me a nice 4TB WD RE drive for Christmas. I haven't had a chance to put it in. It's a SAS and I don't have any devices that take 3.5" SAS. Too late to return it so I can either, buy a SATA for ~$200 or buy a used PowerEdge R710 with 48 GB RAM and 8 cores for ~$240. Guess which one is on it's way....

      posted in Water Closet
      stacksofplatesS
      stacksofplates
    • RE: Chrome Browser in Ubuntu 17.04

      @thwr said in Chrome Browser in Ubuntu 17.04:

      @stacksofplates said in Chrome Browser in Ubuntu 17.04:

      @travisdh1 said in Chrome Browser in Ubuntu 17.04:

      @Lakshmana said in Chrome Browser in Ubuntu 17.04:

      @scottalanmiller Full DVD means all iso in the ubuntu page?

      After a quick glance at the alternate downloads for Ubuntu, it doesn't appear that they have an "Everything DVD" like Debian provides.

      Ya you would probably have to make a local repo for everything. Never done it on Ubuntu, but it's relatively easy with RHEL.

      That's what I thought too and probably the best possible approach to keep an offline Linux system updated. It's not even specific to Debian or Ubuntu, you can basically do the same with Windows (e.g. WSUS).

      This should get you started:
      https://help.ubuntu.com/community/AptGet/Offline/Repository
      https://help.ubuntu.com/community/Repositories/Personal
      http://askubuntu.com/questions/170348/how-to-create-a-local-apt-repository
      http://linoxide.com/ubuntu-how-to/setup-local-repository-ubuntu/

      PS: Follow one tutorial and adapt information from the other links later on, if required.

      Ya I that with all of our RHEL systems. Yum gives you a reposync tool, so I just reposync the repos I'm attached to and store them on an apache server. It's really easy.

      posted in IT Discussion
      stacksofplatesS
      stacksofplates
    • RE: What Are You Doing Right Now

      @scottalanmiller said in What Are You Doing Right Now:

      @stacksofplates said in What Are You Doing Right Now:

      Wow, Oracle sucks. It requires you to have a user called "oracle" (which is fine) but it also requires a separate group to be the primary GID. So rather than just using a secondary group like everything else, you now have to have a secondary group that is the primary group. And apparently you have to install it with a GUI. Ridiculous.

      Yup, having used Oracle a lot in the "enterprise", it was never considered to be a very serious product. It's a bit of a joke.

      Ya luckily never really had to deal with it before this. For the price it should install RHEL on a hypervisor and configure itself in the VM and make me coffee while I wait.

      posted in Water Closet
      stacksofplatesS
      stacksofplates
    • RE: Seperate VM's or one to rule them all?

      We keep all of the services separate. If they are light services you could run a VM as a container host for LXC or Docker.

      posted in IT Discussion
      stacksofplatesS
      stacksofplates
    • RE: What Are You Doing Right Now

      Finished my Bind role for Ansible. Now I don't need to back up my DNS server any more.

      posted in Water Closet
      stacksofplatesS
      stacksofplates
    • RE: NextCloud LDAP Error

      @JaredBusch said in NextCloud LDAP Error:

      @scottalanmiller said in NextCloud LDAP Error:

      @Dashrender said in NextCloud LDAP Error:

      @scottalanmiller said in NextCloud LDAP Error:

      @wirestyle22 said in NextCloud LDAP Error:

      @scottalanmiller said in NextCloud LDAP Error:

      @wirestyle22 said in NextCloud LDAP Error:

      @scottalanmiller said in NextCloud LDAP Error:

      @wirestyle22 said in NextCloud LDAP Error:

      @scottalanmiller said in NextCloud LDAP Error:

      Start with... what is the base of your AD?

      CN=<username>,CN=FCC,CN=Divison of Information Technology,CN=administration,CN=Departments,DC=domaincontroller.domain,DC=org

      So when you join a Windows desktop to the domain, you put domaincontroller.domain.org?

      That's not normal.

      I tried it both ways 😞

      Before we try things, let's be systematic. What do you put onto Windows machines to do this? Ignore NextCloud for the moment.

      The last time I've ever had to do anything with LDAP was like 7 years ago working at the hospital. It's been a really long time for me.

      So you are joining NextCloud to AD that you've never even joined a Windows desktop to? What's the NAME of your AD Domain?

      Right.

      Let's assume we are talking about NTG here. They domainname might be ntg.co so when adding a computer, you would type in ntg.co, not servername.ntg.co

      Or more commonly, because that would be a blunder that NTG would not make, it would be something like ad.ntg.co as the domain and a DC would be something like ny-win-dc1.ad.ntg.co.

      Correct the current Microsoft recommended standard is ad.yourrealdomain.tld

      The old standard of domain.local has been not a standard for years and years.

      So my test AD infrastructure (because we do not use AD for anything real) is ad.bundystl.com and the DC is bundydc01.ad.bundystl.com

      I do the same even for my Linux stuff. My house is pa.jhbcomputers.com

      posted in IT Discussion
      stacksofplatesS
      stacksofplates
    • RE: Fitness and Weightloss

      @scottalanmiller said in Fitness and Weightloss:

      I switched to using a standing desk tonight. So maybe this will help me stay more active all throughout the day in general. The only seat that I have for this desk is broken, so no way to sit down at all right now.

      Ha that's less of "switched to" and more just "I'm not able to sit at my desk". 😛

      posted in Water Closet
      stacksofplatesS
      stacksofplates
    • Ansible Syntax

      So I recently learned that you can use a YAML dictionary for your tasks instead of the normal key=value syntax. So for example this task:

      - name: Copy issue template
        template: src=issue.j2 dest=/etc/issue owner=root group=root mode=0644
      

      is the same as:

      - name: Copy issue template
        template:
          src: issue.j2
          dest: /etc/issue
          owner: root
          group: root
          mode: 0644
      

      At least to me the second way is easier to read, and is more in line with other CM tools like Puppet where that would be:

      file { '/etc/issue':
          content => template('ssh/issue.erb'),
          owner   => root,
          group   => root,
          mode    => 644,
      }
      
      posted in IT Discussion ansible yaml configuration managment linux
      stacksofplatesS
      stacksofplates
    • RE: What Are You Doing Right Now

      In VxRail training.

      posted in Water Closet
      stacksofplatesS
      stacksofplates
    • RE: I moved to Linux!

      @aaronstuder said in I moved to Linux!:

      @scottalanmiller Have you used boxes? Very impressive.....

      Use VirtManager. Boxes is way limited. I can control everything with VirtManager running on my Chromebook.

      • Atom
      • Gimp
      • Inkscape
      • Audacious
      • Backintime
      • Remmina
      • Cockpit
      • Tmux
      • Ansible
      • Chrome (for Netflix and such)
      • VLC
      • Pithos

      No particular order here, just stuff I have.

      If you really want to be focused, use i3 instead of a full DE.

      posted in IT Discussion
      stacksofplatesS
      stacksofplates
    • RE: Something not said enough;

      @Romo said in Something not said enough;:

      @gjacobse said in Something not said enough;:

      Thank you - I feel that I have continued to learn from input from a number of people...

      @scottalanmiller
      @JaredBusch
      @travisdh1
      @DustinB3403

      and a number of other people...

      So - I express my thanks.

      I would like to jump on the thank you train and extend my gratitude for those four as well and would like to include @stacksofplates. You're posts are super helpful and insightful =). Thank you, everybody, for all the knowledge and contributions!!

      Of course I would also like to thank you @gjacobse for all the help you have provided with all NTG related things as well 😃

      Aw shucks.

      posted in Water Closet
      stacksofplatesS
      stacksofplates
    • RE: netdata 1.5 released - big update!

      @scottalanmiller said in netdata 1.5 released - big update!:

      How do I see my servers taht are not on the Internet, for example? Let's say I have 1,000 servers, how do I view them? The purpose of a central console is so that I have one place, one secured place, to go view them. If each machine has its own dashboard, I have to get every one of them out on the Internet so that I can view them?

      I can't even think of 1 server (that I have) where I would this level and speed of real time data let alone 1,000. I know they are out there but I know I don't have any. SNMP and collectors give me way more info than I can use and allow for central monitoring. This seems like a super niche product that could have security implications.

      posted in IT Discussion
      stacksofplatesS
      stacksofplates
    • RE: Collision Domain - In POS

      @scottalanmiller said in Collision Domain - In POS:

      @WrCombs said in Collision Domain - In POS:

      @scottalanmiller said in Collision Domain - In POS:

      @WrCombs said in Collision Domain - In POS:

      @scottalanmiller said in Collision Domain - In POS:

      @WrCombs said in Collision Domain - In POS:

      So it's not a collision Domain because their Obsolete in wired networks due to switches -

      I'm not sure what else could be causing the issue other than bad cabling/ Bad terminals/ Bad switches.
      May be worth replacing the other bar switch and changing out the Cabling under the bar.

      So start with the basics and see what you can track down. If you are getting congestion or bad switches / cabling, you'd expect to see it even with a ping. Ping from device to device and see if you ever get latency or drops. If not, likely it's not a networking thing.

      Can you describe the freezing? Like if you do X which devices freeze. And if you do Y, these other devices freeze.

      Only information I have is " when we're Busy, and we try to order items on the terminals they run slow and freeze up. then it comes back then it shows up. the delay is anywhere from 15 seconds to 2 minutes"
      Nothing specific when it comes to what their doing on the terminals.

      Any reason that there is a network suspicion then, why not suspect the server?

      Cause it's the exact same image/setup as the other site they have, and the other site doesn't run slow what so ever.

      Still, many times more likely to be a hard drive or something.

      This. Something like that trans.log is being written to a single 5400 RPM drive.

      posted in Water Closet
      stacksofplatesS
      stacksofplates
    • 1
    • 2
    • 9
    • 10
    • 11
    • 12
    • 13
    • 143
    • 144
    • 11 / 144