Eric B.'s questions

I've got a dual Xeon X5650 (6core) HP ProLiant DL160 G6 with 72G RAM and an HP P410 4port hardware raid card that I want to repurpose as a NAS and a test server. Additionally, I have 4 x 4T drives in it.

My initial plan was to just HW RAID5 the 4 drives and install Citrix Xenserver 7.2 (free version) on it. I have a couple of Windows VMs that I nede to run, as well as a PFSense firewall. But the primary use would be to create a main VM to run Docker (CoreOS or CentOS or Ubuntu), and a large 8T Virtual Drive assigned to that VM. Then run different Docker images mounted to different volumes created within that large VHD.

Some of the docker images I would like to run:

• Plex
• Sonarr
• Gitlab
• Docker-timemachine
• nextCloud
• ...

Part of the 8T virtual drive would also be exposed as a NAS for sharing space in the office (this is for a small business context).

So now, I'm wondering if my approach is sound at all; should I be putting this much faith in hardware raid vs just using ZFS across the 4 drives instead (and dropping the HW RAID altogether)?

One concern I have of the HW RAID is if the controller fails. Then I end up have 12T of data that can only be recovered by another P410 controller (and I don't have a hot spare).

On the other hand, I am having difficulty finding docs if/how to install Citrix Xenserver on ZFS. Everything I've found seems to discuss ZFS as a Storage Repo for XS, but nothing relating to install DOM0 as a ZFS install. Secondly, I do have concerns of any ZFS management / maintenance required through DOM0. Normally, I like to have DOM0 run as much stock as possible - this simplifies upgrades.

Is installing Citrix XenServer on a ZFS even feasible? ie: the Dom0 would be installed and running on a 12T ZFS. Given that XenServer is based on CentOS7, I figure it should be doable, but not sure how. But even if it is feasible, is it recommended?

Is running XenServer on ZFS (if doable) a safe and recomended alternative to running a HW Raid5 (with FBWC)?

Can someone clarify the difference between ActiveMQ and ActiveMQ-Artemis? From what I understood on Wikipedia, ActiveMQ-Artemis is essentially the successor to ActiveMQ, but if that is the case, I do not understand why Artemis is listed as a subproject of ActiveMQ.

Moreover, if I look at the versioning for ActiveMQ and ActiveMQ-Artemis, I see that the versions are completely different. ActiveMQ being at 5.15 and Artemis being at 2.2.

From what I understand, Artemis is a faster, more recent implementation, including support for JMS 2.0. Given a new installation & deployment of an MQ platform, is should ActiveMQ 5.15 be considered as deprecated in lieu of Artemis?

I'm having some trouble with a Bitbucket -> Jenkins plugin, and would like to see the exact API request that Jenkins is receiving.

I figure this should be fairly easy, but haven't been able to find the correct loggers to enable. I wanted to create a new Log Recorder, but do not know which loggers to enable to log the full incoming requests.

How can I enable request logging in Jenkins so I can see the full/exact request that is being made?

I'm having trouble extending my partition on a XenServer 6.5. I originally had a 1.4TiB device (hardware RAID5) that I since upgraded to 10.9TiB. I upgraded the RAID by swapping out one disk at a time until the 4 disks were replaced.

I then used the RAID configuration to extend the RAID to the full 10.9TiB size.

=> ctrl slot=4 ld all show status
   logicaldrive 1 (10.9 TB, 5): OK

I've next told Xen dom to rescan the scsi device:

echo 1 >  /sys/block/sda/device/rescan 

Now when I start gdisk, I see the following:

Disk /dev/sda: 23441913520 sectors, 10.9 TiB
Logical sector size: 512 bytes
Disk identifier (GUID): A28B730F-3064-494E-BD7E-DDE8CAD5A12D
Partition table holds up to 128 entries
First usable sector is 34, last usable sector is 2930122766
Partitions will be aligned on 2048-sector boundaries
Total free space is 6042 sectors (3.0 MiB)

Number  Start (sector)    End (sector)  Size       Code  Name
   1            2048         8388641   4.0 GiB     0700  
   2         8390656        16777249   4.0 GiB     0700  
   3        16779264      2930122766   1.4 TiB     8E00  

So I see that gdisk is able to see the full 10.9TiB, but I can't seem to delete and recreate parititon 3 to be larger than the original 1.4TiB. Furthermore, it says last usable sector is 2930122766.

How can I get gdisk to allow me to create a partition that goes from sectors 16779264-23441913520?

I've recently installed the hpasmcli on my ProLiant DL160 G6 and wanted to see what kind of power draw it was having. However, when I try to use the show powermeter command, I get a "no power meters detected".

hpasmcli> show powermeter
No Power Meters were detected.

hpasmcli> show powersupply
Power supply #1
    Present  : Yes
    Redundant: No
    Condition: Ok
    Hotplug  : Not supported

My unit only has a single power supply in it, so I understand this output.

hpasmcli> show server
System        : ProLiant DL160 G6  
Serial No.    : USEXXXXX6L
ROM version   : O33 07/01/2013
iLo present   : No

Does the "no power meters detected" mean that my powersupply doesn't support that feature? Is that what the iLo present : No means? Is there a way to upgrade my power supply model to one that includes a power meter? Is there a way to see the model number of the powersupply currently in my HP?

I'm trying a new install of XenServer 6.2 on an HP Proliant DL160G6 with a HP P410 raid card configured as RAID 5 over 4 drives, with a 256M BBWC. The install starts off properly, but when it gets to about 56% completion of "Installing from Base Pack", it just completely freezes and stops continuing. I am unable to switch consoles using Alt-F2, etc.

I've tried rebooting and restarting the install, but to no additional success.

If, however, I reconfigure my HW Raid as Raid 1 or Raid 1+0, the installation is successful, which makes me wonder if this is a compatibility issue with P410 Raid configuration, however, according to the HCL, the P410 is listed as compatible.

I've tried running Inquisitor as a stress test on the RAID, and it passes flawlessly in RAID-5 configuration.

How can I further diagnose where the problem lies? Is there a way to see exactly where/why this is failing? What would cause it to fail on Raid 5 but not Raid 1?

I'm trying to setup SNMP traps on my HP DL160 G6 (Xenserver) to notify me whenever I have issues with the server.

Following suggestions from this serverfault post, I've installed the HP SNMP agents and configured my cma.conf to email me whenever I have an SNMP trap generated. To test it out, I unplugged a secondary NIC.

My problem is that instead of just getting a single trap indicating that the NIC was disconnected or reconnected, I get traps generated every 10secs indicating that the link is down or up.

I also installed the HP System Management web interface, but can't seem to find how to configure the NIC agent to stop sending traps every 10secs and only trap on state change.

Is this possible?

With the SNMP agent (cmanicd) enabled, I get the following in /var/log/messages:

Feb 28 16:51:01 xenserver-opvorsfu logger: 192.168.0.152: Enterprise Specific Trap (18012)
Feb 28 16:51:01 xenserver-opvorsfu logger:  Uptime: 4:32:23.51
Feb 28 16:51:01 xenserver-opvorsfu logger: SNMPv2-MIB::sysName.0 = STRING: xenserver-opvorsfu
Feb 28 16:51:01 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.11.2.11.1.0 = INTEGER: 0
Feb 28 16:51:01 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.18.2.3.1.1.5.2 = INTEGER: 0
Feb 28 16:51:01 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.18.2.3.1.1.10.2 = INTEGER: 2
Feb 28 16:51:01 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.2.2.4.17.0 = STRING: "Not Available"
Feb 28 16:51:01 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.18.2.3.1.1.14.2 = INTEGER: 4
Feb 28 16:51:01 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.1.2.13.1.1.5.5.0 = STRING: "HP NC362i Integrated DP Gigabit Server Adapter"
Feb 28 16:51:01 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.18.2.3.1.1.32.2 = ""
Feb 28 16:51:01 xenserver-opvorsfu logger: IP-MIB::ipAdEntAddr.0.0.0.0 = IpAddress: 0.0.0.0
Feb 28 16:51:01 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.18.2.4.1.1.6.2 = STRING: "N/A"
Feb 28 16:51:12 xenserver-opvorsfu cmanicd: Entering iml_log_link_down(slot: 0, port: 2) 
Feb 28 16:51:12 xenserver-opvorsfu cmanicd: Entering log_iml_event(slot: 0, port: 2, code: (Down,2)) 
Feb 28 16:51:12 xenserver-opvorsfu cmanicd: Entering get_event_id(slot: 0, port: 2 
Feb 28 16:51:12 xenserver-opvorsfu cmanicd: No Existing event id found for the slot and port. 
Feb 28 16:51:12 xenserver-opvorsfu cmanicd: an IML entry for this slot and port does not exist, so add one. 
Feb 28 16:51:12 xenserver-opvorsfu cmanicd: cmanic: could not log IML event, retval = 4 errno = 0, logevt.c,774  
Feb 28 16:51:12 xenserver-opvorsfu logger: Feb 28 16:51:01 xenserver-opvorsfu logger: Feb 28 16:50:51 xenserver-opvorsfu logger: Feb 28 16:50:40 xenserver-o
pvorsfu logger: Feb 28 16:50:40 xenserver-opvorsfu logger: Feb 28 16:50:39 xenserver-opvorsfu logger: Feb 28 16:50:29 xenserver-opvorsfu logger: Feb 28 16:5
0:19 xenserver-opvorsfu logger: Feb 28 16:50:08 xenserver-opvorsfu logger: Feb 28 16:49:58 xenserver-opvorsfu logger: Feb 28 16:49:57 xenserver-opvorsfu log
ger: Feb 28 16:49:54 xenserver-opvorsfu kernel: [16281.101268] igb: eth1 NIC Link is Down
Feb 28 16:51:12 xenserver-opvorsfu logger: 192.168.0.152: Enterprise Specific Trap (18012)
Feb 28 16:51:12 xenserver-opvorsfu logger:  Uptime: 4:32:33.98
Feb 28 16:51:12 xenserver-opvorsfu logger: SNMPv2-MIB::sysName.0 = STRING: xenserver-opvorsfu
Feb 28 16:51:12 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.11.2.11.1.0 = INTEGER: 0
Feb 28 16:51:12 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.18.2.3.1.1.5.2 = INTEGER: 0
Feb 28 16:51:12 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.18.2.3.1.1.10.2 = INTEGER: 2
Feb 28 16:51:12 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.2.2.4.17.0 = STRING: "Not Available"
Feb 28 16:51:12 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.18.2.3.1.1.14.2 = INTEGER: 4
Feb 28 16:51:12 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.1.2.13.1.1.5.5.0 = STRING: "HP NC362i Integrated DP Gigabit Server Adapter"
Feb 28 16:51:12 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.18.2.3.1.1.32.2 = ""
Feb 28 16:51:12 xenserver-opvorsfu logger: IP-MIB::ipAdEntAddr.0.0.0.0 = IpAddress: 0.0.0.0
Feb 28 16:51:12 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.18.2.4.1.1.6.2 = STRING: "N/A"
Feb 28 16:51:15 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:52919 
Feb 28 16:51:15 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:32864 
Feb 28 16:51:15 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:51237 
Feb 28 16:51:22 xenserver-opvorsfu cmanicd: Entering iml_log_link_down(slot: 0, port: 2) 
Feb 28 16:51:22 xenserver-opvorsfu cmanicd: Entering log_iml_event(slot: 0, port: 2, code: (Down,2)) 
Feb 28 16:51:22 xenserver-opvorsfu cmanicd: Entering get_event_id(slot: 0, port: 2 
Feb 28 16:51:22 xenserver-opvorsfu cmanicd: No Existing event id found for the slot and port. 
Feb 28 16:51:22 xenserver-opvorsfu cmanicd: an IML entry for this slot and port does not exist, so add one. 
Feb 28 16:51:22 xenserver-opvorsfu cmanicd: cmanic: could not log IML event, retval = 4 errno = 0, logevt.c,774  
Feb 28 16:51:22 xenserver-opvorsfu logger: Feb 28 16:51:12 xenserver-opvorsfu logger: Feb 28 16:51:01 xenserver-opvorsfu logger: Feb 28 16:50:51 xenserver-opvorsfu logger: Feb 28 16:50:40 xenserver-opvorsfu logger: Feb 28 16:50:40 xenserver-opvorsfu logger: Feb 28 16:50:39 xenserver-opvorsfu logger: Feb 28 16:50:29 xenserver-opvorsfu logger: Feb 28 16:50:19 xenserver-opvorsfu logger: Feb 28 16:50:08 xenserver-opvorsfu logger: Feb 28 16:49:58 xenserver-opvorsfu logger: Feb 28 16:49:57 xenserver-opvorsfu logger: Feb 28 16:49:54 xenserver-opvorsfu kernel: [16281.101268] igb: eth1 NIC Link is Down
Feb 28 16:51:22 xenserver-opvorsfu logger: 192.168.0.152: Enterprise Specific Trap (18012)
Feb 28 16:51:22 xenserver-opvorsfu logger:  Uptime: 4:32:44.43
Feb 28 16:51:22 xenserver-opvorsfu logger: SNMPv2-MIB::sysName.0 = STRING: xenserver-opvorsfu
Feb 28 16:51:22 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.11.2.11.1.0 = INTEGER: 0
Feb 28 16:51:22 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.18.2.3.1.1.5.2 = INTEGER: 0
Feb 28 16:51:22 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.18.2.3.1.1.10.2 = INTEGER: 2
Feb 28 16:51:22 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.2.2.4.17.0 = STRING: "Not Available"
Feb 28 16:51:22 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.18.2.3.1.1.14.2 = INTEGER: 4
Feb 28 16:51:22 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.1.2.13.1.1.5.5.0 = STRING: "HP NC362i Integrated DP Gigabit Server Adapter"
Feb 28 16:51:22 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.18.2.3.1.1.32.2 = ""
Feb 28 16:51:22 xenserver-opvorsfu logger: IP-MIB::ipAdEntAddr.0.0.0.0 = IpAddress: 0.0.0.0
Feb 28 16:51:22 xenserver-opvorsfu logger: SNMPv2-SMI::enterprises.232.18.2.4.1.1.6.2 = STRING: "N/A"
Feb 28 16:51:30 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:41836 
Feb 28 16:51:30 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:50204 
Feb 28 16:51:30 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:47990 
Feb 28 16:51:32 xenserver-opvorsfu cmanicd: Entering iml_log_link_down(slot: 0, port: 2) 

With cmanicd disabled, I get the following in /var/log/messages:

Feb 28 16:44:28 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:55759 
Feb 28 16:44:28 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:33561 
Feb 28 16:44:28 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:48321 
Feb 28 16:44:32 xenserver-opvorsfu kernel: [15959.771627] igb: eth1 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
Feb 28 16:44:43 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:52043 
Feb 28 16:44:43 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:52082 
Feb 28 16:44:43 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:46917 
Feb 28 16:44:58 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:42179 
Feb 28 16:44:58 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:56542 
Feb 28 16:44:58 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:52481 
Feb 28 16:45:09 xenserver-opvorsfu kernel: [15996.671268] igb: eth1 NIC Link is Down
Feb 28 16:45:14 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:58277 
Feb 28 16:45:14 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:57163 
Feb 28 16:45:14 xenserver-opvorsfu snmpd[19168]: Received SNMP packet(s) from UDP: [127.0.0.1]:58464 

I've got an internal bind server which works fine for my local name resolution. Most of the hosts are mapped with internal IPs instead of their corresponding external IPs. However, there are situations in which there are no equivalent internal IPs. These hosts are currently only listed in my public DNS server. Consequently, when I try to query the internal DNS server, I get unknown hosts.

Is there anyway I can configure BIND to query an external source if it receives a request for a host that it does not know, even though this host is in the domain for which it is authoritative?

For example, if my domain is called: site.com

Internal DNS may have:

www.site.com    192.168.1.1
smtp.site.com   192.168.1.2
mail.site.com   192.168.1.3

External DNS may have:

www.site.com    199.200.201.1
smtp.site.com   199.200.201.2
mail.site.com   199.200.201.3
ftp.site.com    199.200.201.4
support.site.com 199.200.201.5

If I query internal DNS for www.site.com, it will return 192.168.1.1. But if I query it for ftp.site.com, I'll get an unknown host. Is there anyway to configure my internal bind to forward that request to the external server and return 199.200.201.4?

Or is my only choice to copy the missing hosts to my internal DNS server, and have two copies of the same records (one on my public DNS and one on my internal DNS)?

I'm having trouble with DNS configuration for a domain and do not know how to trace it properly. From trying to troubleshoot with nslookup, it seems as though the authoritative NS are never being contacted. But when I check the whois for the domain, they are correctly listed.

I've tried to enable the debug output for nslookup, but that does not particularly help. The Autority Records are listed as empty.

Is there a "trace" mode that I can use to see the full queries? I'd like to see the full steps that an nslookup is making. Ie: query against current NS server, NS server contact registrar for listed autoritative NS, then query against Authoritative NS.

I'm not entirely sure where my configuration hiccup is. I believe that the NS servers listed at the registrar are CNAMEs. Are NS listed in the registrar allowed to be CNAME records, or must they be A records?

I'm doing development for a webapp that sends emails out to users. I've typically used Argosoft Mail Server (Freeware) in a windows environment as an SMTP server in which I've enabled relay but shutdown actual delivery. This allow me to point my webapp to my local server, and am able to validate all emails that are sent out by looking at the outgoing queue. Additionally, to be 100% safe, I would put in a non-existant DNS server to use to ensure that if I ever accidentally enable delivery, the server would still not be able to look up any external addresses and deliver the mail.

Now that I am doing development on an OSX machine, I'm looking for a similar solution. By default, OSX comes with Postfix installed, so I figured I would just use that and configure it similarly. However, I cannot seem to find an option anywhere to disable all outgoing email. Essentially, I want to set up Postfix as an open relay server, but disable delivery of all email. I can then look at the outgoing queue and see if the email(s) are as expected.

Is there anyway to easily accomplish this? Or is there a better solution I can use?