Managing Power
Understanding How Dynamic Power Management Enables Better Utilization of Power
You can use the dynamic power management feature to better utilize the power available in the power entry module (PEM). Whether or not a new hardware component is powered on depends on the availability of power in the PEM. A component is not powered on if the PEM cannot meet the worst-case power requirement for that component. Starting in Junos OS Release 15.1R1, MX Series routers support dynamic power management. Starting in Junos OS Release 17.2R1, EX9200 switches support dynamic power management.
The maximum power that each type of MIC consumes is maintained in a static database. The chassis
daemon process (chassisd), which manages power budgeting for all line
cards, uses this data when budgeting power for MICs. MICs are brought online only after
the chassis daemon verifies that the worst-case power required for the MICs and the
power required for all the online FRUs (Field Replaceable Units: Replaceable or
swappable Junos device and device parts ) are available in the PEM.
In Junos OS Release
15.1R1, for MX Series routers, dynamic power management for MICs is
disabled by default. You can enable the feature
by enabling the mic-aware-power-management statement at
the [edit chassis] hierarchy level. When dynamic power
management is disabled, the chassis daemon checks for the worst-case
power requirement of the MPC and the MICs before allocating power
for the MPC. Whereas, when mic-aware-power-management statement
is enabled, the chassis daemon considers the power requirement of
only the MPCs. The worst-case power consumption by the MICs is not
considered while the chassis daemon budgets power for the MPC. Power
budgeting for MICs is done only after the MPC is powered on and the
MICs come online. Every time you disable or enable dynamic power management,
you must restart the chassis or the MPC for the changes to take effect.
In Junos OS Release 17.2R1, for EX9200 switches, dynamic power management for MICs is enabled by default.
Starting from Junos OS Release 17.3R1, for MX10003 routers, mic-aware dynamic power management is enabled by default.
Starting from Junos OS Release 18.2R1, for JNP10K-LC2101 MPC on MX10008 routers, dynamic power management is enabled by default. However, dynamic power management for MICs is not supported on JNP10K-LC2101 because JNP10K-LC2101 is a fixed configuration MPC and supports only built-in PICs.
After you enable the dynamic power management feature, use the set chassis
preserve-fpc-poweron-sequence configuration mode command to preserve the
sequence in which MPCs are powered on. This configuration is required to maintain the
order in which the MPCs come online after a router or switch restart.
In Junos OS Release 15.1F5 and later, dynamic power management is enabled by default on several MPCs. Models include MPC3E-3D-NG, MPC3E-3D-NG-Q, MPC2E-3D-NG, MPC2E-3D-NG-Q, MPC6E, MPC7E-MRATE, and MPC7E-10G on MX240, MX480, MX960, MX2010, and MX2020 and on MPC8E and MPC9E on MX2010, and MX2020 Universal Routing Platforms.
Understanding Power Management on the PTX5000
Starting in Junos
OS Release 14.1, the power management feature for PTX5000 routers
ensures that at any time, the chassis power requirements do not exceed
the available chassis power. The PTX5000 has
two PDUs to meet the power requirements of the chassis. Each PDU is
capable of providing power to the chassis on its own. In case the
power requirement exceeds the individual capacity of a PDU, the required
power is provided by both the PDUs and the No redundant
power supply alarm is triggered. If the system cannot
provide power for all the installed FPCs or PICs, the system brings
down FPCs or PICs that in can no longer provide power for and the Insufficient Power - FRU(s) went offline alarm is
raised.
The power management feature provides the following functionality:
Power management ensures that high-priority FPCs continue to receive power when the system does not have sufficient power to keep all the FPCs online.
Power management ensures that if a power supply fails, the router can continue to operate normally by keeping high-priority FPCs online and taking low-priority FPCs offline.
If power supply failure requires power management to power down some components, power management does so by gracefully powering down lower-priority FPCs.
Power management manages power to router components by employing a power budget policy. In its power budget policy, power management:
Budgets power for each installed router component that requires power. The amount that power management budgets for each component is the maximum power that component might consume under worst-case operating conditions. For example, for the fan tray, power management budgets the amount of power required to run the fans at their maximum speed setting, even if the current fan speed is much lower.
Manages the router for N+N power redundancy, which ensures uninterrupted system operation if one power supply fails.
Provides power to host subsystem components, such as the Routing Engines, before it provides power to the FPCs.
Manages the priority of individual FPCs. By assigning different priorities to the FPCs, you can determine which FPCs are more likely to receive power in the event of insufficient power.
Power Priority of FPCs
The power priority of FPCs determines:
The order in which FPCs are allocated power.
How power is reallocated if there is a change in power availability or demand in an operating router.
This section covers:
- How an FPC’s Power Priority Is Determined
- FPC Priority and FPC Power Allocation
- FPC Priority and Changes in the Power Budget
How an FPC’s Power Priority Is Determined
Using the CLI, you can assign an explicit power priority to an FPC slot. The power priority is determined by the slot number, with the lowest-numbered slots receiving power first. Thus, if you do not explicitly assign priorities to slots, power priority is determined by slot number, with slot 0 having the highest priority. See Configuring Power-On Sequence to Redistribute the Available Power.
FPC Priority and FPC Power Allocation
When a PTX5000 is powered on, power management allocates power to components according to its power budget policy. After power management has allocated power to the host subsystem components, it allocates the remaining available power to the FPCs. It powers on the FPCs in the configured order of priority until all FPCs are powered on or the available power provided by both the PDUs is exhausted. Thus if available power is exhausted before all FPCs receive power, higher-priority FPCs are powered on while lower-priority FPCs remain powered off.
FPCs that have been taken offline are not allocated power.
Because power management does not allocate power to an FPC that
has been taken offline, that FPC is brought online only when you commit
a configuration. You must explicitly use the request chassis
fpc slot slot-number online command to
bring an FPC online that was taken offline previously.
If an FPC with a high priority in the priority sequence also has high-power requirement, and if the system does not have the required power available, then the lower priority FPCs with lower power requirements are also not powered on. This is to maintain consistency and also avoid powering off of the lower priority FPC when extra power is available. For example, if an FPC that requires 450 W has a higher priority than an FPC that requires 330 W, then the FPC with the lower power requirement (330 W) is also not powered on if the system does not have the required power to power the FPC that requires 450 W.
FPC Priority and Changes in the Power Budget
In an operating router, power management dynamically reallocates power in response to changes in power availability or demand or changes in FPC priority. Power management uses the configured priority on FPC slots to determine how to reallocate power in response to the following events:
When a new power supply is brought online, FPCs that were powered off because of insufficient power are powered on in the order of priority.
When a user changes the assigned power priority of one or more FPCs when power is insufficient to meet the power budget, power management reruns the current power budget policy and powers FPCs on or off based on their priority. As a result, FPCs receive power strictly by the order of priority and previously operating FPCs might no longer receive power.
When an FPC is installed, Junos OS does not automatically power on and bring the FPC online. This FPC stays in the offline state until the user brings it online through the CLI or by pushing the online button, and only if the available chassis power is more than the budgeted power for this FPC, the FPC becomes operational.
Power Zones
In a PTX5000 equipped with high capacity PDUs and PSMs, there in one common zone that provides power to all FRUs and all FPCs. A high-capacity PDU can support up to eight PSMs and it does not support power zoning, unlike a normal-capacity PDU. All available PDU power is considered as a part of single zone. All PSMs provide power to the common zone. The PSM LEDs on the craft interface are interpreted as described in PTX5000 Craft Interface LEDs. After the PDU upgrade from the normal-capacity PDUs to High-Capacity PDUs, the power management converges all power zones into a single common zone. All FRU power is distributed based on the power available in the common zone.
Presence of both normal-capacity PDUs and high-capacity PDUs is referred to as mixed-mode of operation and is supported only during the PDU upgrade.
To cater for the increase in the PIC power consumption, the power manager is enhanced to account for the PIC power separately from the FPC. The priority sequence for the PICs follows the priority sequence for the FPCs. That is, PICs installed in high-priority FPCs are given preference over PICs installed in low-priority FPCs. All PICs on an FPC have the same priority.
You cannot mix existing PDUs with the High Capacity DC PDU.
Power Supply Redundancy
By default, power management in PTX5000 routers is configured to manage the power supplies for N+N redundancy, by which power supplies are held in reserve for backup if the other power supplies are removed or fail.
When power is insufficient to meet the budgeted power requirements, power management raises alarms as follows:
With power supply redundancy, when one PSM fails, it does not cause FPCs to go offline. Only the
No redundant power supply alarmis raised. However, with no redundancy, FPCs can go offline depending on the total chassis power available at that time. When an FPC or PIC goes offline due to insufficient power, which is indicated byNo powerin the output of theshow chassis fpccommand, then theInsufficient Power - FRU(s) went offlinealarm is raised. The alarm gets cleared when there is sufficient power to bring up all the FPCs and PICs. TheInsufficient Power - FRU(s) went offlinealarm is raised when PSMs fail, when PSMs are powered off manually, or any time there is insufficient power for the system to power all the FPCs or PICs in the system.When power fails or when a PSM is removed, power management:
Calculates the total chassis power available from the remaining PSMs for the FPCs.
Powers off the FPCs based on the priority depending on the power budget for the FPCs and the FRUs and their configured power-on sequence.
Note:In the scenario where the available power is more than the budgeted power required by the FPC but less than its maximum power, the FPC is taken offline and then brought online, but one or more PICs in that FPC are not online.
When a new PSM is inserted, power management:
Checks the power-on sequence of the FPCs and the PICs and brings any offline PICs online when power is available.
Powers on the FPCs based on the FPC’s budgeted power and its power-on sequence depending on its priority.
Maintains the power for high-priority FPCs and their PICs by taking the low-priority FPCs offline when all the FPCs are brought online, depending on the available power.
Power management clears all alarms when sufficient power is available to meet normal operating and reserved power requirements.
Power Redundancy on SRX5400
The power redundancy feature in SRX5400 supports to manage the high-capacity high line power supplies for 2+2 AC redundancy mode. The power rate is 1167W at low line and 2050W at high line on SRX5400. The 2+2 redundancy mode requires four AC power supplies.
The minimum PSU requirement is now 2 instead of 1 for the PEM alarm to be raised. If you install only 1 high-capacity high line AC, a major alarm is raised.
For more information about power supply on SRX5400 refer to SRX5400 Services Gateway AC Power Supply Specifications.
T4000 Power Management Overview
Starting with Junos OS Release 12.3, the power management feature is enabled on a Juniper Networks T4000 Core Router. This feature enables you to limit the overall chassis output power consumption. That is, this feature enables you to limit the router from powering on a Flexible PIC Concentrator (FPC) when sufficient output power is not available to power on the FPC during booting or normal operation.
The power management feature is enabled only when six input feeds with 40 amperes (A) each or four input feeds with 60 A each is configured on the router. The power management feature is not enabled for any other input feed–-current combination. When the power management feature is not enabled, Junos OS tries to power on all the FPCs connected to the router.
: If you do not configure the power management feature and the maximum power draw is exceeded by the router during booting or normal operation, FPCs’ states might change from Online to Offline or Present, some traffic might drop, or the interfaces might flap.
Interface flapping occurs when a router alternately announces the state of the interface to be as up and down in quick sequence.
After you connect the input feeds to the router, you must configure
the number of input feeds connected to the router and the amount of
current received at the input feeds. Use the feeds statement
and the input current statement at the [edit chassis
pem] hierarchy level to configure the number of input feeds
and the amount of current received at each input feeds, respectively.
You can connect three 80 A DC power cables to the
six-input DC power supply by using terminal jumpers. When you do this,
ensure that you configure the feeds statement to have the
value 6 and the input current statement to have
the value 40. If these configurations are not set, the
power management feature is not enabled and,
therefore, Junos OS tries to power on all the FPCs connected to the
router.
When the power management feature is enabled, FPCs connected to the router are powered on based on the power received by the router. If the router receives sufficient power to power on all the FPCs connected to the router, all the FPCs are powered on. If sufficient power is not available, Junos OS limits the number of FPCs brought online. That is, Junos OS uses the total available chassis output power as a factor to decide whether or not to power on an FPC connected to the router.
Of all the supported FPCs of a T4000 router, the T1600 Enhanced Scaling FPC4 (model number: T1600-FPC4-ES) has the greatest power requirement.Table 1 compares the FPC connection limits between a six-input feed 40 A connection and a four-input feed 60 A connection when power management is enabled and T1600-FPC4-ES is connected to router.
Six Input Feeds with 40 A Connection |
Four Input Feeds with 60 A Connection |
|---|---|
When T1600-FPC4-ES is not connected:
|
When T1600-FPC4-ES is not connected:
|
When only one T1600-FPC4-ES is connected:
|
When only one T1600-FPC4-ES is connected:
|
When only T1600-FPC4-ES FPCs are connected:
|
More than one T1600-FPC4-ES cannot be brought online. |
When the power management feature is enabled, FPC power-on consistency is not maintained across router reboots. That is, the same set of FPCs that were powered on before a reboot might not be powered on after the reboot. Before the router reboot, the FPCs are powered on according to their insertion order in the chassis. After the reboot, the FPCs are powered on according to the FRU power-on sequence configured in the
fru-poweron-sequencestatement at the[edit chassis]hierarchy level. If the FRU power-on sequence is not configured, Junos OS uses the ascending order of the slot numbers of the FPCs as the sequence to power on the FPCs.Removal of any online FPC from the chassis does not change the state of any other FPC and does not trigger the power management feature to power on the FPCs that were not powered on initially because of the lack of sufficient power. When any online FPC is removed from the chassis, if you need to trigger the power management feature to re-evaluate the situation, you need to reboot or restart the chassis. Alternatively, you can make a configuration change at the
[edit chassis]hierarchy level and then issue thecommitcommand to commit the changes made at the[edit chassis]hierarchy level. The power management feature to re-evaluates the situation when a configuration change is committed at the[edit chassis]hierarchy level.
Configuring the Six-Input DC Power Supply on T Series Routers
By default, the six-input DC power supply is configured
to have all the six input feeds connected. You can also choose to
provide four or five input feeds to the six-input DC power supply.
When providing four or five input feeds on standalone routers, you
need to configure the feeds statement at the [edit
chassis pem] hierarchy level. When providing four or five input
feeds to an LCC router in a routing matrix, you need to configure
the feeds statement at the [edit chassis lcc lcc-number pem] hierarchy level.
Starting with Junos OS Release 12.3, the power management feature
is enabled on T4000 routers with six-input DC power supply. The power
management feature is enabled only when six input feeds with 40 amperes
(A) each or four input feeds with 60 A each is configured on
the router. To do this, you need to configure the feeds and input-current statements at the [edit chassis
pem] hierarchy level.
Before configuring input feeds for your router, see the T640 Core Router Hardware Guide, T1600 Core Router Hardware Guide, or T4000 Core Router Hardware Guide for special considerations and for the number of input feeds supported by the router.
The value assigned to the
feedsstatement must be equal to the number of input feeds provided to the power supply. Else, an alarm message is generated to indicate the mismatch.
The following procedures describe how to configure the six-input DC power supply on different routers:
- Configuring the Six-Input DC Power Supply on an LCC Router in a Routing Matrix
- Configuring the Six-Input DC Power Supply on T640 and T1600 Routers
- Configuring the Six-Input DC Power Supply on T4000 Routers
Configuring the Six-Input DC Power Supply on an LCC Router in a Routing Matrix
To configure the six-input DC power supply on an LCC router in a routing matrix:
Configuring the Six-Input DC Power Supply on T640 and T1600 Routers
To configure the six-input DC power supply on a standalone T640 or T1600 router:
Configuring the Six-Input DC Power Supply on T4000 Routers
To configure the six-input DC power supply on a T4000 router:
Redistributing the Available Power by Configuring Power-On Sequence
Routers running on Junos OS Release 10.0 and later support an enhanced AC Power Entry Module (PEM) to provide the necessary power infrastructure to support up to twelve higher-capacity DPCs with higher port density and slot capacity. To support the cooling requirements for the enhanced AC PEMs, the routers support enhanced fan trays and fans.
The default behavior for MPC power-on sequence is slot number based, that is, slot 0 is brought online first followed by slot 1, slot 2 up to slot 11. For the scenarios, where it is running a mix of high capacity line cards (for core facing), and low capacity line cards (for access facing) in their system, you can use the fru-poweron-sequence option to manually set the MPC power on sequence and hence ensure that the more important core facing line cards are brought online first irrespective of which slots these are in. This approach provides fine control over deterministically bringing up MPCs, howeverConfiguring Power-On Sequence to Redistribute the Available Powerr, it is heavy on configuration and entails to follow the discipline in slot to MPC mapping across all the systems.
The Junos OS enables you to configure the power-on sequence for the DPCs on an MX Series router chassis containing the new AC PEM. This enables you to redistribute the available power to the DPCs based on your requirements and the calculated power consumption of the DPCs. To configure the power-on sequence, refer to related information.
See Also
Configuring Power-On Sequence to Redistribute the Available Power
You can configure the power-on sequence for the Flexible PIC Concentrators (FPCs) on MX, PTX, and T routers. This configuration enables you to redistribute the available power to the FPCs on the basis of your requirements and the calculated power consumption of the FPCs.
To configure the power-on sequence:
If the configured sequence contains invalid numbers, Junos OS considers only the valid numbers in the sequence. The invalid numbers are silently discarded.
If the power-on sequence is not configured by including the
fru-poweron-sequencestatement, Junos OS uses the ascending order of the slot numbers of the FPCs as the sequence to power on the FPCs.Issue the show chassis power command to view power limits and usage details for the FPCs.
See Also
Configuring Voltage Level Monitoring of FPCs
You can monitor the voltage on the flexible PIC concentrator (FPC) at regular intervals. When the voltage falls below 10%, the FPC is offlined.
The faulty FPC is monitored at 500ms intervals. The output of
the show chassis fpc command shows Power
Failure for the faulty FPC. The FPC remains in powered
down state until the voltage level is normal again.
Enabling Voltage Failure Errors on the FPC
fpc-nmi-volt-fail-knob controls the behavior
of the FPC after detecting voltage failure, and to online or offline
the FPC based on the voltage level. To enable monitoring the voltage
level on the FPC:
Navigate to the
[edit chassis]hierarchy level.Include the
set chassis fpc-nmi-volt-fail-knob enablestatement to enable voltage monitoring on the FPC.[edit chassis] { fpc-nmi-volt-fail-knob enable; }
Disabling Voltage Failure Errors on the FPC
To disable monitoring the voltage level on the FPC:
Navigate to the
[edit chassis]hierarchy level.Include the
set chassis fpc-nmi-volt-fail-knob disablestatement to disable voltage monitoring on the FPC.[edit chassis] { fpc-nmi-volt-fail-knob disable; }
Overriding the Default Maximum Power (Junos OS Evolved)
On the PTX10001-36MR router, you can override the maximum power value of the power supply module (PSM) by specifying a lesser power value. Similarly, on the PTX10008 router, you can override the default power budget allocated to the line card by specifying a power value.
- Overriding the Default Maximum Power (PTX10001-36MR)
- Overriding the Default Maximum Power (PTX10008)
Overriding the Default Maximum Power (PTX10001-36MR)
You can override the maximum power value of a power supply module
(PSM), if you need to deploy the PTX10001-36MR router in an environment
that does not require the maximum power capacity (3000 W) of the PSM.
You can use the command set chassis psm max-power to override
the maximum power capacity of the PSM. Using this configuration, you
can specify a value that is less than the maximum capacity of the
PSM, and then monitor the real-time power consumption against the
configured power value.
See the following example to know how to override the default power in PTX10001-36MR:
user@router#set chassis psm max-power 1600user@router#commit
If the above configuration is set, the system power capacity
is shown as 1600W. See the following show chassis power detail output:
user@router# show chassis power detail
Chassis Power Voltage(V) Power(W)
Total Input Power 937
PSM 0
Input 1 229 391
Output 12.03 305.44
Capacity 1600 W (maximum 3000 W)
PSM 1
Input 1 0 546
Output 12.04 515.08
Capacity 1600 W (maximum 3000 W)
Item Used(W)
Routing Engine 0 25
CB 0 5
System:
Zone 0:
Capacity: 3200 W (maximum 6000 W)
Actual usage: 937 W
Total system capacity: 3200 W (maximum 6000 W)
If the power consumption of the PTX10001-36MR router exceeds
the threshold you configured using the set chassis psm max-power command, the software does not take any corrective action against
the breach; and the router might still encounter a power failure.
If the power consumption exceeds the configured threshold, the system raises a chassis alarm, as shown in the following example:
user@router# show system alarms Mar 15 12:51:30 2 alarms currently active Alarm time Class Description 2020-03-15 12:50:52 UTC Minor Power consumption is critical
Overriding the Default Maximum Power (PTX10008)
On the PTX10008 router, during the system startup, the power
management software by default takes the maximum power mentioned for
each field replaceable unit (FRU) and makes the power calculations
based on this number. However, you can override the default power
budget allocated to the line card by specifying a power value (in
watts). You can use the command set chassis fpc fpc-slot max-power watts to override the default
power. You can use the command show chassis fpc detail to
view the maximum power consumption by a line card.
You can also disable the power management on PTX10008 by using
the command set chassis no-power-budget. If you disable
the power management on PTX10008, the system does not move any of
the FRUs to offline state in case of insufficient power. Instead,
the system keeps all the FRUs powered on by default. However, in case
of a power shortage, a power redundancy alarm is raised as shown in
the following example.
user@router> show system alarms 1 alarm currently active Alarm time Class Description 2019-07-25 21:16:25 UTC Major chassis No Redundant Powe
See Also
Powering Off Packet Forwarding Engines
You can power on or power off the Packet Forwarding Engines in a running system, or keep a Packet Forwarding Engine powered off when the FPC comes online. The following are a couple of scenarios in which this feature is used.
- When the Packet Forwarding Engine ASIC is malfunctioning.
- To conserve power in case the deployment does not require the full capacity of the system.
To power off a Packet Forwarding Engine, use the following steps:
user@host# set chassis fpc slot-number pfe
pfe-id power on
user@host# commit
You need to apply this configuration to both the Packet Forwarding Engines in an ASIC to be able to commit the configuration.
On MX series routers with MPC10E-15C-MRATE, you can power off or power on only the Packet
Forwarding Engine 2. The Packet Forwarding Engines 0 and 1 do not support this command. On
the MPC10E-15C-MRATE, operating the Packet Forwarding Engine 2 requires the Packet
Forwarding Engines 0 and 1 to be functional. You can use the command show chassis
fpc fpc-lot detail to view the Packet Forwarding Engine power
ON/OFF status and bandwidth for the individual Packet Forwarding Engines in the
MPC10E-15C-MRATE.
You can use the show chassis fpc fpc-slot detail command to view the Packet Forwarding Engine power on/off configuration status. See an example below:
user@router> show chassis fpc 0 detail
Slot 0 information:
State Online
Temperature 41 degrees C / 105 degrees F (PFE_24-HBM)
Temperature 44 degrees C / 111 degrees F (PFE_25-HBM)
Temperature 43 degrees C / 109 degrees F (PFE_26-HBM)
Temperature 41 degrees C / 105 degrees F (PFE_27-HBM)
Temperature 40 degrees C / 104 degrees F (PFE_28-HBM)
Temperature 40 degrees C / 104 degrees F (PFE_29-HBM)
Temperature 38 degrees C / 100 degrees F (PFE_30-HBM)
Temperature 39 degrees C / 102 degrees F (PFE_31-HBM)
Start time 2020-10-28 00:46:17 PDT
Uptime 1 day, 1 hour, 34 minutes, 48 seconds
Max power consumption 825 Watts
PFE Information:
PFE Power ON/OFF Bandwidth SLC
0 On 500
1 On 500
2 On 500
3 On 500
4 On 500
5 On 500
6 On 500
7 On 500
Power Saving Mode (ACX7100-48L, ACX7100-32CD)
In addition to the normal mode, we provide support to enable or disable the
Power Saving mode on ACX7100-48L and ACX7100-32CD. In power saving mode,
you can save power upto 48 W. Please note:
- After configuring this feature, you must reboot your system for the changes to take effect.
- You can configure this feature only on the following ports:
-
On ACX7100-48L: et-0/0/24 to et-0/0/47 and et-0/0/51 to et-0/0/53
-
On ACX7100-32CD: et-0/0/16 to et-0/0/31 and et-0/0/34 to et-0/0/35
-
Enabling Power Saving Mode
-
If PTP (Precision Time Protocol) is enabled on the last port (47 or 31 depending on the device) in the member-range, you cannot configure the power saving feature on the unused interface range. The sample configuration varies for each platform. If you try to configure this feature on an unused port with PTP enabled, the system will display an error message.
- The ports mentioned in the example is for ACX7100-48L.
Disabling Power Saving Mode
Use Feature Explorer to confirm platform and release support for specific features.
See Also
Power Mode Management on PTX10002-36QDD
PTX10002-36QDD device supports two power supply units (PSUs) with 1 + 1 PSU redundancy. The operating mode of PTX10002-36QDD depends on the type (3000 W or 2200 W) of PSUs present in the system. Two 3000 W PSUs are required for the system to operate in normal power mode with 1 + 1 PSU redundancy. If one of the PSUs is absent, the system does not support 1 + 1 redundancy and determines the operating mode based on the available PSU. When you use a 2200 W PSU, the system operates in the low power mode.
You can also force the system to operate in low power mode when 3000 W PSUs are present, by using the following CLI command:
set chassis mode power-optimized
You must reboot the system for the mode change to take effect.
This configuration reduces the power consumption of the device. See the following
show chassis power command output:
user@host> show chassis power
Chassis Power Voltage(V) Power(W)
Total Input Power 817
PSM 0
State: Online
Input 1 51 444
Output 12.03 427.17
PSM 1
State: Online
Input 1 51 373
Output 12.02 329.3
System:
Power mode: power-optimized
Zone 0:
Capacity: 6000 W (maximum 6000 W)
Allocated power: 2150 W (3850 W remaining)
Actual usage: 817 W
Total system capacity: 6000 W (maximum 6000 W)
Total remaining power: 3850 WPower Redundancy for JNP10K-PWR-AC3 Power Supply Module
The JNP10K-PWR-AC3 power supply provides N+1 PSM (Power Supply Module) redundancy support for the MX10004, MX10008, PTX10004 and PTX10008 platforms. The JNP10K-PWR-AC3 PSM is equivalent to four power-supplies that consist of four input feeds (A0, A1, B0, and B1) with a maximum power output capacity of 7.8 KW. You can enable redundancy at the source or feed level.
The redundancy feature makes the system more reliable. It enables the system to raise an alarm[1] when the power remaining in the system is less than the power of the connected PSM with the highest capacity.
When a PSM is removed or has a fault in a redundancy-disabled system, the power manager will power off the required FRUs.
See hardware guide for more information about JNP10K-PWR-AC3 PSM.
JNP10K-PWR-AC3 PSMs cannot provide PSM redundancy in a 4 slot chassis with high power consuming line cards (LCs) such as JNP10K-LC1301.
The configuration examples and alarms given below are for the MX10004 and MX10008 platforms. On
the PTX10004 and PTX10008 platforms, use psm instead of
pem when configuring the redundancy.
Source Redundancy
When you have two power sources (source A and source B), you can enable source redundancy by connecting two sets of independent power feeds from each source to the four input terminals of JNP10K-PWR-AC3 PSM.
Please note the following conditions before configuring the source redundancy:
The software currently supports only one redundant source: for example, source A (main source) along with source B (backup or redundant source).
The source redundancy is disabled by default. You can use the following CLI command to enable source redundancy:
set chassis pem redundancy source-redundancy
Source redundancy is only applicable if all the PSMs in the system are JNP10K-PWR-AC3 PSMs. If you try to include a JNP10K-PWR-AC2 PSM in the power supply, the system will raise an alarm[2].
You must ensure that the feed distribution is even.
It is not possible to enable source redundancy and feed redundancy simultaneously. Therefore, you need to disable feed redundancy before enabling source redundancy. To disable feed redundancy, use the following CLI command:
delete chassis pem redundancy feed-redundancy slot pem slot number
-
You must set the DIP switches to the
feed expectedposition for all four PSM input feeds; otherwise, the system will raise an alarm[3].
Please note the following conditions after configuring the source redundancy feature:
The Power Manager considers the capacity of each JNP10K-PWR-AC3 PSM as a 2-feed capacity. See hardware guide for more information.
If one of the sources becomes unavailable, the system will raise an alarm[4] corresponding to the failed feeds. In addition, the system will raise another alarm[3] and disable the source redundancy until the source failure is fixed.
The overall system power capacity reduces to protect against a source failure.
The software will simulate the feature and determine the new system power capacity. If the system’s new power capacity cannot accommodate the existing system load, the system will disable the source redundancy, raise an alarm[5], and continue to operate in normal mode.
When a PSM is faulty and there is no source failure, the system capacity is expected to reduce further. If the reduced system capacity cannot support the existing load, the system will disable the source redundancy, raise an alarm[5], and continue to operate in normal mode. You should replace the failed PSM as soon as possible.
Feed Redundancy
You can enable feed rudundancy by connecting all the four input feeds (A0, A1, B0, and B1) of JNP10K-PWR-AC3 PSM to one or more power sources. When one feed is down, the other feed will continue providing the power and keep the platform operational.
Feed redundancy is not supported for JNP10K-PWR-AC2 PSMs and JNP10K-PWR-BLN3
active blank modules. On a Junos OS device, if you try to configure feed
redundancy for these unsupported devices, the system will ignore the
configuration and make a print available in the LCMD logs. On a Junos OS Evolved
device, the system will make the Feed redundancy unsupported for
PSM print available in the messages.
Please note the following conditions before configuring the feed redundancy:
-
The PSM must have at least two feeds connected.
-
The software currently supports only one redundant feed.
-
You must set the DIP switches to the
feed expectedposition for both primary and redundant feeds. Otherwise, the system will raise an alarm[6]. Additionally, you must confirm that the DIP switch configuration matches the connected feeds. -
Feed redundancy is disabled by default. You can use the following CLI command to enable feed redundancy:
set chassis pem redundancy feed-redundancy slot pem slot number
-
It is not possible to enable source redundancy and feed redundancy simultaneously. Therefore, you need to disable source redundancy before enabling feed redundancy. To disable source redundancy, use the following CLI command:
delete chassis pem redundancy source-redundancy
Please note the following conditions after configuring the feed redundancy feature:
-
The Power Manager will calculate the capacity of the JNP10K-PWR-AC3 PSM by subtracting one feed from the total number of connected feeds. See hardware guide for more information.
-
If the redundant feed becomes unavailable, the system will raise an alarm[4] corresponding to the failed feed. In addition, the system will raise another alarm[7] and disable feed redundancy until the feed failure is fixed.
-
The overall system power capacity reduces to protect against a feed failure.
-
The software will simulate the redundancy feature and determine the system's new power capacity. If the system’s new power capacity is not sufficient to accommodate the existing system load, the system will disable the feed redundancy on a Junos OS device. However, on a Junos OS Evolved device, the system will disable feed redundancy sequentially on PSMs until it achieves the required capacity. Both operating systems will also raise an alarm[8] and continue to operate in normal mode.
-
When a PSM is faulty and there is no feed failure, the system capacity is expected to reduce further. If the reduced system capacity is not sufficient to support the existing load, the system will disable the feed redundancy, raise an alarm[8], and continue to operate in normal mode. You should replace the failed PSM as soon as possible.
|
Sl No |
Message displayed in the output of |
Message displayed in the output of |
Description |
|---|---|---|---|
|
1 |
|
|
Appears when the power remaining in the system is less than the individual connected PSMs. |
|
2 |
|
|
Appears when you connect an unsupported power supply. |
|
3 |
|
|
Appears when you do not position the DIP switches for all feeds. |
|
4 |
|
|
Appears when a source or redundant feed is not available. |
|
5 |
|
|
Appears when the power is not sufficient after enabling the source redundancy. |
|
6 |
|
|
Appears when you do not position the DIP switches for all connected feeds. |
|
7 |
|
|
Appears when the feed redundancy is disabled. |
|
8 |
|
|
Appears when the power is not sufficient after enabling the feed redundancy. |
Change History Table
Feature support is determined by the platform and release you are using. Use Feature Explorer to determine if a feature is supported on your platform.