fan reversed on asw1-ulsfo
Closed, ResolvedPublic
Actions

Assigned To

Authored By

	• Gage
	Jun 6 2014, 7:56 PM

Description

The replacement PSU is onsite @ ulsfo in unitedlayer storage.

12:50 < jgage> asw1-ulsfo's LCD says "LED:SPD ALARM 00". that mean anything to you guys? i haven't looked it up yet.
12:52 < paravoid> 2014-02-14 20:50:32 UTC Major FPC 2 Airflow direction of PSU and Chassis fan are not matching
12:52 < paravoid> Power Supply 0 REV 01 740-044237 EK0712366608 PS 1000W AC FORWARD AIR FLOW CONTROL
12:52 < paravoid> Power Supply 1 REV 02 740-029666 EM0710461561 PS 1000W AC REVERSE AIR FLOW CONTROL
12:52 < paravoid> Fan Tray Fan Tray, Back to front Airflow

Referred To By:
T83992: Juniper monitoring

Details

Reference: rt7642

Related Objects

Mentioned In: T83992: Juniper monitoring
Mentioned Here: T83992: Juniper monitoring

Event Timeline

• rtimport raised the priority of this task from to Medium.Dec 18 2014, 1:55 AM

• rtimport added a project: ops-ulsfo.

• rtimport set Reference to rt7642.

• Gage created this task.Jun 6 2014, 7:56 PM

-- According to the juniper forum. The power supplies we are using are have
mismatched airflow.
http://forums.juniper.net/t5/Junos/Airflow-direction-mismatch-alarm/td-p/205961
Chris Johnson
Operations Engineer
Wikimedia Foundation, Inc
(415) 578-0844
<cmjohnson at wikimedia>

Status changed from 'new' to 'open' by RT_System

Reference by ticket #7654 added by dzahn

The same alarm text showed on the LCDs of both switches in the stack, asw1 and
asw2. I'm not certain which chassis actually has the problem.

Dependency on ticket #8596 added by robh

I've created a ticket and linked it as blocker for the ordering of a
replacement power supply.
--
Rob Halsell
Operations Engineer
Wikimedia Foundation, Inc.
E-Mail: <rhalsell at wikimedia>

Part numbers for front-to-back vs back-to-front cooling:
http://www.juniper.net/documentation/en_US/release-independent/junos/topics/reference/specifications/cooling-system-ex4500.html

The order of this PSU is being tracked in RT8596, which has been escalated for ops mgmt approval before order.

RobH changed the task status from Open to Stalled.Dec 29 2014, 4:26 PM

RobH assigned this task to • Gage.

RobH changed the visibility from "WMF-NDA (Project)" to "Public (No Login Required)".

RobH set Security to None.

• Gage updated the task description. (Show Details)Dec 30 2014, 7:35 PM

@Gage, I know you went on-site on Friday; can you update this with the status?

Restricted Application added a project: acl*sre-team. · View Herald TranscriptJan 30 2015, 9:42 AM

Thanks for the reminder. Onsite visit was Weds, 2015-01-28. I updated the procurement ticket in https://rt.wikimedia.org/Ticket/Display.html?id=8596 but forgot to also update Phab:

Wrong part was received. We received 'coupagjeta' which blows front to back, but the switches are mounted with their backs to the cold row so we actually need another part 'coupae8eaa' which blows back to front.

Erroneous part is in SF office awaiting return shipment.

faidon mentioned this in T83992: Juniper monitoring.Feb 11 2015, 10:16 AM

In T83978#1004360, @Gage wrote:

Erroneous part is in SF office awaiting return shipment.

This is RT #8596.

RobH updated the task description. (Show Details)Mar 13 2015, 7:38 PM

@mark: Can you or @faidon (or possibly @BBlack) confirm availability on Tuesday?

Per IRC conversation with @mark, someone with in depth juniper knowledge should be around during this, in the event of unexpected consequences.

Ideally (for being in Pacific timezone), I'd drive down post rush hour and plan to be onsite @ ulsfo at 10AM Pacific. This would be 17:00 GMT. If everything goes as expected, the swap is a 5 minute process.

If one of you could confirm availability, I'd appreciate it! I've also added this window to the operations maint calendar and invited all three of you (though only one needs to attend.)

Additionally, if no one is available at that time, please suggest alternatives. Please note that anything from 8:30-9:30AM Pacific (15:30-16:30 GMT) is rush hour commuting here, and hard to hit without driving to the datacenter much earlier (in which case an earlier window is just easier than that particular hour window.)

I have the paperwork to ship the mis-shipped unit (currently locked in my desk @ office) back tomorrow. I'm resolving this onsite ticket, since the power supply has been swapped.

fan reversed on asw1-ulsfoClosed, ResolvedPublicActions

Description

Details

Related Objects

Event Timeline

fan reversed on asw1-ulsfo
Closed, ResolvedPublic
Actions