Brocade ICX Series (cheap & powerful 10gbE/40gbE switching)

[Dave Corder]

Maybe the onboard voltage regulators that step down 52V to 12V are dead

Thanks, that should be fairly easy to find and test with a multimeter.

 

[andrea87]

Thanks, that should be fairly easy to find and test with a multimeter.

Could you take a good resolution pic of the main board of the switch? We might be able to give you some more information on where to check for voltages.

The only internal pic I can find is a low resolution 640x480 one.

If you check for voltages after the inductor on the center-left of the pic, do you get anything? That looks to be an insulated main 52 to 12V switcher. Then there are two more converters by the looks of it in the top right part of the board, which could probably be 5/3.3/1.8V supplies. If you get no voltage on any of those rails, also check for shorts (with the switch off) on any of the main mosfets for these rails (soic-8 for the main supply and smaller devices for the others)

 

[RoachedCoach]

I've got my ICX7250-24P I bought off eBay in Europe. I believe someone already did a fan swap to it, as I've found two delta FFB0412VHN spliced onto the original connectors.

I've tried running the quieter fans I've bought (EFB0412VHD-00), but the asic temp gets way too hot. The airflow at "slow" speed / 4.5V feels at least half of what the fast FFB fans do move. This is a show chassis run after 15 mins of power up, with no interfaces up besides one ethernet for management:

SSH@ICX7250-24P Router>show chassis
The stack unit 1 chassis info:

Power supply 1 (AC - PoE) present, status ok
Power supply 2 not present

Fan 1 ok, speed (auto): [[1]]<->2
Fan 2 ok, speed (auto): [[1]]<->2

Fan controlled temperature:
        Rule 1/2 (MGMT THERMAL PLANE): 85.6 deg-C
        Rule 2/2 (AIR OUTLET NEAR PSU): 39.5 deg-C

Fan speed switching temperature thresholds:
        Rule 1/2 (MGMT THERMAL PLANE):
                Speed 1: NM<-----> 93       deg-C
                Speed 2:        82<----->105 deg-C (shutdown)
        Rule 2/2 (AIR OUTLET NEAR PSU):
                Speed 1: NM<-----> 58       deg-C
                Speed 2:        49<----->105 deg-C (shutdown)

Fan 1 Air Flow Direction:  Front to Back
Fan 2 Air Flow Direction:  Front to Back
Slot 1 Current Temperature: 85.6 deg-C (Sensor 1), 39.5 deg-C (Sensor 2)
Slot 2 Current Temperature: NA
        Warning level.......: 100.0 deg-C
        Shutdown level......: 105.0 deg-C

I've tried adding a baffle made of thin cardboard to move more air through the ASIC's heatsink, but it only made the temperatures worse, hitting even 88°C on the main chip.

Moving back to the faster fans, the switch stabilized at 36-38°C on the PSU and 81-82°C on the main chip. Tomorrow I might try also with a 60x60x10 fan as some of you did on the asic. I'll be at best at 40W pulled out of the PoE ports, if I can get a balance between temps and noise it would be great.

I think one of the issues is that those rear fans you put in there are only 10 CFM. That stock ones were around 25, so it's now getting less than half the original air movement.

When I did my mod and used the Mechantronix MR4020X12B1-RSR I think they do well because they're rated at 17.3 CFM. They are a touch louder than the ones you have now, but very reasonable. I also agree the 60mm fan on the ASIC will likely serve you better.

If for some reason you find it still isn't sufficient, one hacky thing you could do is strap a fan externally next to the PSU inlet on the side. I know someone who has it in a closet that did this. It'll provide more airflow explicitly to the PSU (which it seems is actually the first thing to overheat post-mod).

 

[andrea87]

If for some reason you find it still isn't sufficient, one hacky thing you could do is strap a fan externally next to the PSU inlet on the side. I know someone who has it in a closet that did this. It'll provide more airflow explicitly to the PSU (which it seems is actually the first thing to overheat post-mod).

The main thing with the PSU is that, when running at full power, it's supposed to deliver 370W to the PoE ports plus 50 ish watts to the switch itself. That makes a 400-420W load, that at an efficiency of 85-90% (the psu looks to me to be an active PFC, LLC resonant converter) will push out some 40-60W heat on top of the 50W the switch already dissipates at idle. We're talking above 100W in an 1U form factor with 2 fans, which will require some airflow.

My plan is to run the switch way lower in PoE power, currently I've got three TP-Link omada access points that draw around 5W each, and a couple security cameras / doorbells that do at best 3-4W. If I'm getting to 40W it'll be a lot. I've bought this switch mostly for the 8x10g sfp side and to consolidate 2 switches into a single one.

I like the idea of doing some diy stuff to push air into the side psu port. I've got about 4.5cm space from the switch to the side of the rack cabinet, and the intake vent is 12x3cm. I could easily design and 3d print an adapter to fit two or three small 40x40 fans for some additional airflow.

 

[clcorbin]

I have a question regarding the ICX 7750-48F. I currently have two ICX 6610 running as a stack. Due my "need" for additional 40Gbe ports, I am going to break the stack and replace the 6610 in the rack with the 7750.

Today, I was working on getting the 7750 configured correctly and assigning VLANs to ports (mostly to match 6610 so it will be a "drop in" as much as possible). I noticed that on the 6610, when I setup a vlan trunk, it displayed Tag = Yes and Pvid = N/A. However, setting up the same trunk on the 7750 shows Tag = Yes and Pvid = 99 (in my case, my default VLAN).

Does this indicated that the ports on the 7750 (software version 08.0.95nT203) run in hybrid mode by default? That was my original guess, however when I tried to add that port to my management vlan as untagged, it said the port was already a member of that vlan (per the trunk).

Feels like I am doing something wrong here due to the newer software version in the 7750...

Regards,

Clint

 

[andrea87]

If for some reason you find it still isn't sufficient, one hacky thing you could do is strap a fan externally next to the PSU inlet on the side. I know someone who has it in a closet that did this. It'll provide more airflow explicitly to the PSU (which it seems is actually the first thing to overheat post-mod).

And... Done!

I've designed and 3d printed a bracket to add 3x 40x40 fans to the side of an 1U switch. If anyone wants to print one, the stl files are here.



With the fans running at 5V off an USB port, the additional intake lowered the PSU temperature by 4°C and the SoC temp by about 2-3°C. The switch is inside my rack, with the front door closed, and two 120mm fans blowing out exhaust air up.

Before:
Slot 1 Current Temperature: 72.4 deg-C (Sensor 1), 43.5 deg-C (Sensor 2)
After:
Slot 1 Current Temperature: 69.7 deg-C (Sensor 1), 39.5 deg-C (Sensor 2)

 

[tubs-ffm]

Does anybody know about compatibility issues on PoE devices?

I am using my ICX 7150-C12P powering two devices without issues. When I connect my Snom M900 DECT directly to the switch is does not run. It gets power, it boots, but either the ethernet port on ICX is completely down or no data goes through. When I connect the same device with a separate PoE adapter to the same port at the ICX all is fine. The M900 is 802.af class 2 compatible.

Any possibilities to do do some manual setting on the ICX for the port?

 

[andrea87]

Any possibilities to do do some manual setting on the ICX for the port?

with the device attached, can you do a show interface ethe 1/1/x on the port and paste here the results? These are the ones I get on my 7250 with a PoE AP:

Spoiler: show interface ethe 1/1/x

SSH@ICX7250-24P>show interface ethe 1/1/21
GigabitEthernet1/1/21 is up, line protocol is up
Port up for 1 day(s) 4 hour(s) 54 minute(s) 29 second(s)
Hardware is GigabitEthernet, address is 609c.9f06.0880 (bia 609c.9f06.0894)
Configured speed auto, actual 1Gbit, configured duplex fdx, actual fdx
Configured mdi mode AUTO, actual MDI
EEE Feature Disabled
Tagged member of 3 L2 VLANs, untagged in VLAN 1, port state is FORWARDING
BPDU guard is Disabled, ROOT protect is Disabled, Designated protect is Disabled
Link Error Dampening is Disabled
STP configured to ON, priority is level0, mac-learning is enabled
MACsec is Disabled
Openflow is Disabled, Openflow Hybrid mode is Disabled, Flow Control is config enabled, oper enabled, negotiation disabled
Mirror disabled, Monitor disabled
Mac-notification is disabled
VLAN-Mapping is disabled
Not member of any active trunks
Not member of any configured trunks
Port name is AP Esterno
IPG MII 96 bits-time, IPG GMII 96 bits-time
MTU 10200 bytes, encapsulation ethernet
MMU Mode is Store-and-forward
300 second input rate: 19608 bits/sec, 22 packets/sec, 0.00% utilization
300 second output rate: 552456 bits/sec, 51 packets/sec, 0.05% utilization
1736003 packets input, 274338698 bytes, 0 no buffer
Received 1538 broadcasts, 13767 multicasts, 1720698 unicasts
0 input errors, 0 CRC, 0 frame, 0 ignored
0 runts, 0 giants
4324033 packets output, 5622405691 bytes, 0 underruns
Transmitted 152292 broadcasts, 73249 multicasts, 4098492 unicasts
0 output errors, 0 collisions
Relay Agent Information option: Disabled
Protected: No
MAC Port Security: Disabled

From what I can find on the item's page, it's a 10/100 PoE device, you could try forcing the port to work at 100mbit full duplex mode:

(in config mode)
interface ethe 1/1/x
speed-duplex 100-full
exit

the port will cycle, and it will be forced to the desired negotiation speed.

to undo, set speed-duplex to auto on the port.

 

[itronin]

Does anybody know about compatibility issues on PoE devices?

I am using my ICX 7150-C12P powering two devices without issues. When I connect my Snom M900 DECT directly to the switch is does not run. It gets power, it boots, but either the ethernet port on ICX is completely down or no data goes through. When I connect the same device with a separate PoE adapter to the same port at the ICX all is fine. The M900 is 802.af class 2 compatible.

Any possibilities to do do some manual setting on the ICX for the port?

Are you running LLDP on the switch?

 

[tubs-ffm]

Are you running LLDP on the switch?

It is not enabled by intention. I already read that LLDP may cause issues on M900.
Here is my config.

Spoiler: config

SSH@switch01#show run
Current configuration:
!
ver 09.0.10j_cd3T213
!
stack unit 1
  module 1 icx7150-c12-poe-port-management-module
  module 2 icx7150-2-copper-port-2g-module
  module 3 icx7150-2-sfp-plus-port-20g-module
!
!
!
global-stp
!
!
default-vlan-id 2
!
vlan 2 name DEFAULT-VLAN by port
 spanning-tree
!
vlan 10 name DMZ by port
 tagged ethe 1/1/3 ethe 1/2/1
 untagged ethe 1/3/2
 spanning-tree
!
vlan 20 name IoT by port
 tagged ethe 1/1/3 ethe 1/1/5 ethe 1/2/1
 spanning-tree
!
vlan 30 name Guest by port
 tagged ethe 1/1/3 ethe 1/1/5 ethe 1/2/1
 spanning-tree
!
!
!
!
!
!
ip route 0.0.0.0/0 192.168.2.1
optical-monitor
optical-monitor non-ruckus-optic-enable
ipv6 route ::/0 fd00:0:0:2::1
!
!
!
clock summer-time
clock timezone gmt GMT+01
boot system flash primary
jitc enable
hostname switch01
ip dns domain-list domain.lan
ip dns server-address 192.168.2.1 192.168.2.15
ip tftp blocksize 8192
ipv6 dns server-address fd00:0:0:2::1 fd00:0:0:2::15
ipv6 unicast-routing
!
!
ntp
 disable authenticate md5
 master stratum 5
 server 0.de.pool.ntp.org
 server 1.de.pool.ntp.org
 server 2.de.pool.ntp.org
 server 3.de.pool.ntp.org
 peer 192.168.2.1
 peer 192.168.2.3
 peer 192.168.2.10
!
!
!
!
!
!
!
!
!
interface management 1
 ip address 192.168.71.50 255.255.255.0
!
interface ethernet 1/1/1
 port-name OPNsense-LAN
 dhcp6 snooping trust
 dhcp snooping trust
 arp inspection trust
!
interface ethernet 1/1/2
 disable
!
interface ethernet 1/1/3
 port-name opnsense-trunk
 dhcp6 snooping trust
 dhcp snooping trust
 arp inspection trust
!
interface ethernet 1/1/4
 disable
!
interface ethernet 1/1/5
 port-name ap
!
interface ethernet 1/1/6
 disable
!
interface ethernet 1/1/7
 port-name wohnzimmer
!
interface ethernet 1/1/8
 disable
!
interface ethernet 1/1/9
 port-name ip-phone
 no inline power
!
interface ethernet 1/1/10
 port-name NO PoE
 no inline power
!
interface ethernet 1/1/11
 port-name Homeassistant
!
interface ethernet 1/1/12
 disable
!
interface ethernet 1/2/1
 port-name office
 arp inspection trust
!
interface ethernet 1/2/2
 disable
!
interface ve 2
 ip address 192.168.2.2 255.255.255.0
 ipv6 address fd00:0:0:2::2/64
!
!

!
!

ipv6 dhcp6 snooping vlan 2 10 20 30

!
ip dhcp snooping vlan 2 10 20 30
!
ip dhcp snooping verify mac-address
!


!
!
!
!
username super password .....
!
aaa authentication login default local
!
aaa authentication web-server default local
!
aaa authentication snmp-server default local
!
ip dhcp-client disable
!
logging host 192.168.2.4 udp-port 6514
!


!
web-management session-timeout 3600
!
ip ssh encryption aes128-ctr aes192-ctr aes256-ctr


ip ssh key-exchange-method diffie-hellman-group16-sha512 diffie-hellman-group18-sha512 ecdh-sha2-nistp256 ecdh-sha2-nistp384 ecdh-sha2-nistp521

!


!
no telnet server
!
snmp-server community 2 $xxxxx== ro
snmp-server community 2 $xxxx ro
snmp-server location office
snmp-server group admin v3 auth notify all read all write all
snmp-server user homeassistant admin v3 encrypted auth sha xxxx priv encrypted aes xxxx
!
manager disable
manager port-list 987
!
cli timeout 0
!


!


!

end

with the device attached, can you do a show interface ethe 1/1/x on the port and paste here the results? These are the ones I get on my 7250 with a PoE AP:

It is getting confusing. It looks my issue is a different one.

To generate the status for the interface I switched the ports with having the the M900 connected to the PoE adapter. The first the port was up but no I could not get a connection. Switch ports a second time the port changed to status down.

What I noticed are connection errors in the port status and state blocking:

Untagged member of L2 VLAN 2, port state is BLOCKING
[...]
42084 input errors, 33974 CRC, 0 frame, 0 ignored

The issue seems not to be related to PoE as assumed before. Something faulty on the ICX or the M900 I believe.

Status with PoE adapter after 1st port change:

SSH@switch01#show interface ethe 1/1/9
GigabitEthernet1/1/9 is up, line protocol is up
  Port up for 4 minute(s) 20 second(s)
  Hardware is GigabitEthernet, address is d4c1.9e8a.68e0 (bia d4c1.9e8a.68e8)
  Configured speed auto, actual 100Mbit, configured duplex fdx, actual fdx
  Configured mdi mode AUTO, actual MDIX
  Untagged member of L2 VLAN 2, port state is FORWARDING
  BPDU guard is Disabled, ROOT protect is Disabled, Designated protect is Disabled
  Link Error Dampening is Disabled
  STP configured to ON, priority is level0, mac-learning is enabled
  MACsec is Disabled
  Openflow is Disabled, Openflow Hybrid mode is Disabled, Flow Control is config enabled, oper enabled, negotiation disabled
  Mirror disabled, Monitor disabled
  Mac-notification is disabled
  VLAN-Mapping is disabled
  Not member of any active trunks
  Not member of any configured trunks
  Port name is ip-phone
  IPG MII 96 bits-time, IPG GMII 96 bits-time
  MTU 1500 bytes, encapsulation ethernet
  MMU Mode is Store-and-forward
  300 second input rate: 221312 bits/sec, 0 packets/sec, 0.21% utilization
  300 second output rate: 9712 bits/sec, 5 packets/sec, 0.00% utilization
  0 packets input, 7356869 bytes, 0 no buffer
  Received 0 broadcasts, 0 multicasts, 0 unicasts
  27029 input errors, 21849 CRC, 0 frame, 0 ignored
  5079 runts, 101 giants
  1375 packets output, 320704 bytes, 0 underruns
  Transmitted 263 broadcasts, 1110 multicasts, 2 unicasts
  0 output errors, 0 collisions
  Relay Agent Information option: Disabled
  Protected: No
  MAC Port Security: Disabled
 Flexlink: Disabled

 This port is not being monitored for queue drops
Egress queues:
Queue counters    Queued packets        Dropped Packets
         0                 764                   0
         1                   0                   0
         2                   0                   0
         3                   0                   0
         4                 414                   0
         5                  59                   0
         6                   0                   0
         7                 138                   0

Status with PoE adapter after 2nd port change:

SSH@switch01#show interface ethe 1/1/9
GigabitEthernet1/1/9 is down, line protocol is down
  Port down for 1 minute(s) 33 second(s)
  Hardware is GigabitEthernet, address is d4c1.9e8a.68e0 (bia d4c1.9e8a.68e8)
  Configured speed auto, actual unknown, configured duplex fdx, actual unknown
  Configured mdi mode AUTO, actual unknown
  Untagged member of L2 VLAN 2, port state is BLOCKING
  BPDU guard is Disabled, ROOT protect is Disabled, Designated protect is Disabled
  Link Error Dampening is Disabled
  STP configured to ON, priority is level0, mac-learning is enabled
  MACsec is Disabled
  Openflow is Disabled, Openflow Hybrid mode is Disabled, Flow Control is config enabled, oper enabled, negotiation disabled
  Mirror disabled, Monitor disabled
  Mac-notification is disabled
  VLAN-Mapping is disabled
  Not member of any active trunks
  Not member of any configured trunks
  Port name is ip-phone
  IPG MII 96 bits-time, IPG GMII 96 bits-time
  MTU 1500 bytes, encapsulation ethernet
  MMU Mode is Store-and-forward
  300 second input rate: 0 bits/sec, 0 packets/sec, 0.00% utilization
  300 second output rate: 0 bits/sec, 0 packets/sec, 0.00% utilization
  0 packets input, 11371221 bytes, 0 no buffer
  Received 0 broadcasts, 0 multicasts, 0 unicasts
  42084 input errors, 33974 CRC, 0 frame, 0 ignored
  7971 runts, 139 giants
  2125 packets output, 496026 bytes, 0 underruns
  Transmitted 407 broadcasts, 1715 multicasts, 3 unicasts
  0 output errors, 0 collisions
  Relay Agent Information option: Disabled
  Protected: No
  MAC Port Security: Disabled
 Flexlink: Disabled

 This port is not being monitored for queue drops
Egress queues:
Queue counters    Queued packets        Dropped Packets
         0                1200                   0
         1                   0                   0
         2                   0                   0
         3                   0                   0
         4                 634                   0
         5                  76                   0
         6                   0                   0
         7                 215                   0

 

[tubs-ffm]

It is not enabled by intention. I already read that LLDP may cause issues on M900.
Here is my config.

Update.

It looks like I am facing 3 different issues.

1) The ICX has no LLDP but my Ruckus Unleashed. My error is reproducable and the behaviour is as desctribed here. Not related to the ICX 7150
M900 and LLDP protocol - Snom Service Hub - Snom Confluence

I once tested with no lldp enable ports ethe 1/1/9, but the link did not came up.
Maybe this is because of (2)?

2) Still the ICX after reboot sometimes doe not get the port up and I need a second boot. When I connect it to an unmanged switch this indicates only a 10 MBit/s link. So far I tested on ICX to set to 100MBit full duplex because the M900 can do acc. technical spec. Maybe broken?

3) The CRC errors seems to come from a shitty cable or it is related to (2).

 

[RoachedCoach]

And... Done!

I've designed and 3d printed a bracket to add 3x 40x40 fans to the side of an 1U switch. If anyone wants to print one, the stl files are here.

View attachment 38966 View attachment 38967

With the fans running at 5V off an USB port, the additional intake lowered the PSU temperature by 4°C and the SoC temp by about 2-3°C. The switch is inside my rack, with the front door closed, and two 120mm fans blowing out exhaust air up.

Before:
Slot 1 Current Temperature: 72.4 deg-C (Sensor 1), 43.5 deg-C (Sensor 2)
After:
Slot 1 Current Temperature: 69.7 deg-C (Sensor 1), 39.5 deg-C (Sensor 2)

That's quite snazzy and I love it. Good job!

 

[tubs-ffm]

It looks like I am facing 3 different issues.

1) The ICX has no LLDP but my Ruckus Unleashed. My error is reproducable and the behaviour is as desctribed here. Not related to the ICX 7150
M900 and LLDP protocol - Snom Service Hub - Snom Confluence

I once tested with no lldp enable ports ethe 1/1/9, but the link did not came up.
Maybe this is because of (2)?

2) Still the ICX after reboot sometimes doe not get the port up and I need a second boot. When I connect it to an unmanged switch this indicates only a 10 MBit/s link. So far I tested on ICX to set to 100MBit full duplex because the M900 can do acc. technical spec. Maybe broken?

3) The CRC errors seems to come from a shitty cable or it is related to (2).

My issue is "solved", or better to say I found a workaround.

The, in my eyes faulty behaviour of the Snom M900 in combination with LLDP packets, I solved by blocking them on the port of the ICX

no lldp enable ports ethe 1/1/9

The "incompatibility" between both devices not getting a proper link I solved by setting the port to a fixed copper link of 10 MBit.
Acc. spec, the NIC of the M900 is 10/100 MBit. But I only can get 10 Mbit running.

link-config gig copper autoneg-control 10m-full ethe 1/1/9

 

[safado]

Need some help as I'm stuck and not sure where to go to next.

I have a 6610 that is licensed, has the stack removed per Fodeesha etc. I added a 6650 about a year ago and used a 40GB port to connect to the 6610 and it worked perfectly. No issues at all. I had a reboot a couple of days ago and since then I've been unable to get access to the 6650 over the network. I've re-ran setup and assigned the IP, single VLAN, checked licensing and everything else. Wiped back to the default config and started over. I've found that if I connect a 10gb port on the front to a 10gb port of the 6610 it works. The minute I try to use a 40gb qsfp port in the rear (and remove the 10gb port) it will fail. It won't fail by plugging in a qsfp. On the 6650 it shows the port active and state forwarding but on the 6610 it will show it connected and negotiated at 40gb but under state it will show "blocked". Any idea how to get the 6650 and the 6610 chained together (separate switches not a stack) leveraging the 40gb qsfp connection (1/21 or 1/2/6)? Thanks for any help!

 

[DangerMc]

I have a ICX7750 that I've yet to configure. I've updated it to the latest firmware per the fohdeesha docs, but not connected it to any devices yet. If I let it sit turned on, it will periodically reboot (the front display shuts off and then the fans ramp up before they quiet down again), every 10-15 mins or so. I've tried searching to see what the reason for this is but cannot seem to find anything that points to a direct cause. Anyone else experience this with any of their switches?

 

[itronin]

I have a ICX7750 that I've yet to configure. I've updated it to the latest firmware per the fohdeesha docs, but not connected it to any devices yet. If I let it sit turned on, it will periodically reboot (the front display shuts off and then the fans ramp up before they quiet down again), every 10-15 mins or so. I've tried searching to see what the reason for this is but cannot seem to find anything that points to a direct cause. Anyone else experience this with any of their switches?

Is this plugged into a UPS?

 

[Dave Corder]

Is this plugged into a UPS?

Yeah, I had an almost identical issue with my ICX6610-48P when using a consumer-grade CyberPower UPS. I switched it to my rack-mount APC SmartUPS and it's been fine ever since.

 

[andrea87]

The power supplies in these devices are very heavy lifters, they're rated at 1kW output. Also they have to be active PFC due to regulations and such.

as of that, any input waveform that isn't sinusoidal will make the PFC circuit go haywire and possibly trigger very large current spikes on the UPS. It's basically made with a boost converter that charges the "tank" capacitors with an input current that is adjusted to be perfectly sinusoidal. Apply a non-sine input to it and all kinds of mayhem will happen.

I bet your issues are either with a modified sine wave or a square wave ups.

 

[itronin]

I have a ICX7750 that I've yet to configure. I've updated it to the latest firmware per the fohdeesha docs, but not connected it to any devices yet. If I let it sit turned on, it will periodically reboot (the front display shuts off and then the fans ramp up before they quiet down again), every 10-15 mins or so. I've tried searching to see what the reason for this is but cannot seem to find anything that points to a direct cause. Anyone else experience this with any of their switches?

here's a practical recommendation - just plug it into street power and see if it continues the reboot cycle.

 

[DangerMc]

here's a practical recommendation - just plug it into street power and see if it continues the reboot cycle.

Sorry for the late reply, both street power and Eaton 2000VA UPS forces continuous reboot cycles. I have a couple of different UPS' that I will try but never occurred to me that that might be the culprit. (However, I have two different 7450 plugged into the same UPS that does not exhibit the issue)

EDIT : I will try using a different APC SmartUPS and see if that fixes the issue.

Thanks for all the help!