HP StorageWorks 2/128 HP StorageWorks SAN Director 2/128 Fabric OS V4.2.x Rele - Page 22

Issues and Workarounds, not be noticed until the switch is rebooted.

Page 22 highlights

Issues and Workarounds Switch digital certificates are checked when a switch joins a fabric, either because the switch is added to the fabric or because the switch is booting. Changes to the certificate-if, for example, it is removed or corrupted-might not be noticed until the switch is rebooted. Issues and Workarounds Table 6 lists current issues that customers should be aware of and provides workarounds. Table 6: Issues and Workarounds Issue Modifying switch and CP IP addresses can cause a Telnet hang. Changing the switch IP address before changing the CP IP address may cause the CP IP address to become inaccessible on a subsequent attempt to set the switch IP address. After fastbooting the standby CP of the primary FCS and performing a secfcsfailover before HA is in sync, the old primary FCS switch's active CP panics. Downloading configurations by line too quickly on v4.1 switches causes the switch to panic and crash. New switch IP address does not take effect for Web Tools and Fabric Manager in time. If the IP address on a switch is changed, Fabric Manager posts a message that you must restart Fabric Manager to pick up the changes. After restarting Fabric Manager, it does not pick up the IP address change, retaining the old IP address and causing a complete loss of communication to the switch. Workaround When both CP and switch IP addresses need to be changed: set the CP IP address first and then the switch IP address. If this event occurs, telnet into the switch and set the CP IP address again, accepting the default values. When only a switch IP address needs to be changed, set the CP IP address first (keeping the current value) and then the switch IP address to its new value. When only a CP IP address needs to be changed, there is no problem; just change the CP IP address. Issue switchdisable and switchenable to the segmented switch to cause it to rejoin the fabric. Do not execute repeated configdownload commands in rapid succession. After changing a switch IP address, delete the related fabric from Fabric Manager, restart Fabric Manager, and then rediscover the fabric in Fabric Manager. Under heavy stress conditions, wait a few minutes after restarting Fabric Manager for the new switch IP address to take effect. 22 SAN Director 2/128 Fabric OS 4.2.x Release Notes

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24

Issues and Workarounds
22
SAN Director 2/128 Fabric OS 4.2.x Release Notes
Switch digital certificates are checked when a switch joins a fabric, either
because the switch is added to the fabric or because the switch is booting.
Changes to the certificate—if, for example, it is removed or corrupted—might
not be noticed until the switch is rebooted.
Issues and Workarounds
Table 6
lists current issues that customers should be aware of and provides
workarounds.
Table 6:
Issues and Workarounds
Issue
Workaround
Modifying switch and CP IP addresses can cause
a Telnet hang. Changing the switch IP address
before changing the CP IP address may cause the
CP IP address to become inaccessible on a
subsequent attempt to set the switch IP address.
When both CP and switch IP addresses need to
be changed: set the CP IP address first and then
the switch IP address.
If this event occurs, telnet into the switch and set
the CP IP address again, accepting the default
values.
When only a switch IP address needs to be
changed, set the CP IP address first (keeping the
current value) and then the switch IP address to its
new value.
When only a CP IP address needs to be
changed, there is no problem; just change the CP
IP address.
After fastbooting the standby CP of the primary
FCS and performing a secfcsfailover before HA is
in sync, the old primary FCS switch's active CP
panics.
Issue
switchdisable
and
switchenable
to the segmented switch to cause it to rejoin the
fabric.
Downloading configurations by line too quickly
on v4.1 switches causes the switch to panic and
crash.
Do not execute repeated
configdownload
commands in rapid succession.
New switch IP address does not take effect for
Web Tools and Fabric Manager in time.
If the IP address on a switch is changed, Fabric
Manager posts a message that you must restart
Fabric Manager to pick up the changes. After
restarting Fabric Manager, it does not pick up the
IP address change, retaining the old IP address
and causing a complete loss of communication to
the switch.
After changing a switch IP address, delete the
related fabric from Fabric Manager, restart Fabric
Manager, and then rediscover the fabric in Fabric
Manager.
Under heavy stress conditions, wait a few
minutes after restarting Fabric Manager for the
new switch IP address to take effect.