|
|
This chapter covers the tasks required for maintaining a Cache Engine or a cache farm:
If a new release of the Cache Engine system software is available, you can upgrade your system by obtaining the software from Cisco Connection Online (CCO). The URL for the CCO is http://www.cisco.com/cgi-bin/cache-engine.
There are two types of filenames available for download from CCO: files with pax suffix and files with bin suffix. A full-image software is available only by using the pax file; therefore, this is the file you should download. The bin file software can only be used in recovery situations, such as recovering if the Flash memory gets corrupted. Refer to the next section "Recovering the Cache Engine System Software" for instructions on upgrading your software using the bin file.
When you have the software, perform the following steps to upgrade a Cache Engine:
Step 1 Put the upgrade software on your Windows NT or UNIX machine. The upgrade software is one file.
Step 2 From the Windows NT or UNIX machine, download the new image file (for example, ce500-200.pax) to the Cache Engine using FTP.
For example, to download the software using FTP to the Cache Engine 192.168.88.89:
ftp 192.168.88.89
You are placed in the correct directory, which is /local/. If you are not in this directory, use the cd command to change directories:
cd /local/
Step 3 Change to binary mode:
bin
Step 4 Put the file on the Cache Engine. For example, if the upgrade software filename is ce500-200.pax, enter:
put ce500-200.pax
You can now quit the FTP session.
Step 5 Log in to the Cache Engine through the console port.
Step 6 At the Cache Engine console, enter install to install the image. For example, if the filename is ce500-200.pax, enter:
cache# install ce500-200.pax Extracting . Extracting ./imgname.txt Extracting ./etc Extracting ./etc/boot ... (etc) Extracting ./lib/gui/snmp Extracting ./lib/gui/snmp/CISCO-CACHEENGINE-MIB.my Extracting ./lib/gui/snmp/CISCO-SMI.my Extracting ./[image_name].bin.dbg.cus1 File extraction done. ce25.pax: installation is complete. Copy new image '[image_name].bin' to flash memory?[no]: yes Erasing Programming erasing program image sector. 100% complete... You may type [reload] to reboot your box now
Step 7 From the Cache Engine console, enter reload to reboot the Cache Engine:
Console# reload Proceed with reload? [confirm] yes 303949 bytes copying bootagent boot start: 0x7fee000 starting new system - entry 0x8010 0x7dc8800 (tCallback0): NCR810_ Press any key to stop auto-boot...
Step 8 Stop the auto-boot process, by pressing any key, then enter c at the prompt. At the boot device prompt, enter flash. At the filename prompt, press Ctrl-D to quit the bootline. When you see the prompt, enter @.
Press any key to stop auto-boot... [CE500 Series]: c '.' = clear field; '-' = go to previous field; ^D = quit boot device (flash,net) [ net]: flash filename [ ]: ^D . . [CE500 Series]: @
If the Cache Engine's system software becomes corrupted (for example, because of a corrupted Flash memory or a power failure that occurs while you are updating the system software) and you are unable to reload from the Flash memory, you can recover the software using the recovery software available from Cisco Connection Online (CCO). Download the correct file (bin) type.
To recover the Cache Engine system software, perform the following steps:
Step 1 Make sure you have a console connection established. Reboot your Cache Engine manually by toggling the power on/off switch.
Step 2 Stop the auto-boot process by pressing any key.
Enter c to change the auto-boot settings.
Enter net for boot device type.
Enter the appropriate CCO filename, cache IP address, IPO network mask, server IP address, gateway IP address, protocol (ftp or tftp), username, and password.
Press any key to stop auto-boot... 6 [CE500 Series]: c '.' = clear field; '-' = go to previous field; ^D = quit boot device (flash,net) [ flash]: net filename [ ]: /CCO Path/filename.bin cache IP address [ ]: 10.3.1.3 IP network mask [ ]: 255.255.255.0 server IP address [ ]: 171.69.63.242 gateway IP address [ ]: 10.3.1.2 protocol (rcp,ftp,tftp) [ ]: ftp username [ ]: name password [ ]: password flags [ 0]:
Step 3 Enter @ to start the boot sequence.
[CE500 Series]: [CE500 Series]: @ boot device (flash,net) [ net]: filename [/CCO Path/filename.bin]: cache IP address [ 10.3.1.3]: IP network mask [ 255.255.255.0]: server IP address [ 171.69.63.242]: gateway IP address [ 10.3.1.2]: protocol (rcp,ftp,tftp) [ ftp]: username [ name]: password [ password]: flags [ 0]: Loading from the network Attaching network interface fei0... done. Attaching network interface lo0... done. Loading... Starting at 0x108000... .................. Console IO Initialized Attaching network interface lo0... done. Restricted Rights Legend Use, duplication, or disclosure by the Government is subject to restrictions as set forth in subparagraph (c) of the Commercial Computer Software - Restricted Rights clause at FAR sec. 52.227-19 and subparagraph (c) (1) (ii) of the Rights in Technical Data and Computer Software clause at DFARS sec. 252.227-7013. Cisco Systems, Inc. 170 West Tasman Drive San Jose, California 95134-1706 Cisco Cache Engine Copyright (c) 1986-1998 by Cisco Systems, Inc. Software Release: CE ver 1.94 (Build: #26 05/10/1999) Compiled: 18:09:47 May 10 1999 by morlee Image text-base 0x108000, data_base 0x42bea4 System restarted by Power Up The system has been up for 0 seconds. System booted from fei SVCINIT - phase complete Initializing SCSI interface... Initializing SCSI controller... Configuring SCSI peripherals... ID LUN VendorID ProductID Rev. Type Blocks BlkSize pScsiPhysDev -- --- -------- ---------------- ---- ---- -------- ------- ------------ 0 0 QUANTUM VIKING II 9.1SCA 4110 0 17836668 512 0x00fc9458 DEVINIT - phase complete FSINIT - phase complete Configuring from NVRAM... %% Non-volatile configuration memory has not been set up CONF - phase complete NTP: Waiting (2 minutes) for network initialization START - phase complete
Step 4 Reconfigure the Cache Engine with the basic network settings. You can modify the system configuration later using the command-line interface (CLI).
---- System Configuration Dialog ---- You may use Ctrl-D to abort Configuration Dialog at any prompt. Would you like to enter the initial configuration? [yes]: yes host name: domain name: admin password: cache IP address: IP network mask: gateway IP address: DNS server:
If the basic configuration is acceptable, enter yes. The Cache Engine indicates if the configuration is accepted. At this point, the basic configuration is finished.
Use this configuration? [yes]: yes Building configuration... . . START - phase complete
Step 5 Log in to your cache engine with your new admin password. Verify that you are running the correct version by entering show version.
Username: admin Password: test cache# cache# show version Cisco Cache Engine Copyright (c) 1986-1998 by Cisco Systems, Inc. Software Release: Compiled: 17:08:40 Mar 24 1999 by morlee Image text-base 0x108000, data_base 0x419d2c
Step 6 To operate the Cache Management Interface GUI and boot from Flash memory in the future, refer to the "Upgrading the System Software" section in this chapter to download the pax file from CCO and install the pax file.
Step 7 You can now use the Cache Management Interface GUI or the command-line interface (CLI) to enable WCCP or proxy mode and start caching.
The Cache Engine is equipped with one or more disks that are properly partitioned by manufacturing before shipment. For the Model 505, the one disk has dosfs and cache file system (cfs) partitions. For the Model 550, the first disk has dosfs and cfs partitions, and the second disk is partitioned with cfs only. The dosfs partition contains the image file, user interface images, and memory for system output from features such as Syslog.
For the Model 505 model, the disk (disk 0) is located in the center of the Cache Engine. On the Model 550, the first disk (disk 0) is located to the left of the second disk (disk 1).
To verify that a disk is partitioned correctly, execute the show disk partition command. To display the volume names, execute the show disks command.
Console# show disk-partitions /c0t0d0
Disk size : 17836667 sectors
Partition 1: CISCO_UVFAT_1, offset 63 sectors, size 3906250 sectors
Partition 2: CISCO_BFS_1, offset 3906323 sectors, size 1024 sectors
Partition 3: CISCO_CFS_1, offset 3907357 sectors, size 13929300 sectors
Partition 4: UNUSED, offset 0 sectors, size 0 sectors
Console# show disk-partitions /c0t1d0
Disk size : 17836667 sectors
Partition 1: UNUSED, offset 0 sectors, size 0 sectors
Partition 2: CISCO_BFS_1, offset 10 sectors, size 1024 sectors
Partition 3: CISCO_CFS_1, offset 1044 sectors, size 17835613 sectors
Partition 4: UNUSED, offset 0 sectors, size 0 sectors
Console# show disks
/c0t0d0 (scsi bus 0, unit 0, lun 0)
/c0t1d0 (scsi bus 0, unit 1, lun 0)
At any time, you may use the show cfs volumes and show dosfs volumes commands to verify volume status.
Console# show cfs volumes /c0t0d0s3: mounted /c0t1d0s3: mounted Console# show dosfs volumes /c0t0d0s1: mounted /c0t1d0s1: mounted /local: mounted
The dosfs partition may be unmounted if the dosfs unmount command was executed. If the dosfs partition is unmounted, mount the dosfs partition using the dosfs mount command.
If the dosfs mount command does not work properly, execute the show disk-partitions command. Check for the UVFAT partition. The size of the dosfs partition should be approximately 2 gigabytes. A sector is equal to 512 bytes. If there is no disk partition for the dosfs, see to the section "Repartitioning the Cache Engine Disks."
If the dosfs volume does not mount, perform the procedure in the section "Checking and Repairing DOS File System Volumes." If the dosfs volume still does not mount after performing this procedure, repartition and reformat the disks as described in the section "Repartitioning the Cache Engine Disks."
 | Caution Anytime a volume is partitioned or formatted (dosfs or cfs), all the data on the disk is lost. |
| Caution Any time a disk is partitioned, all the data on the disk is lost. |
To repartition the disk 0, which is the location of the dosfs partition, execute the disk partition command.
Console# disk partition /c0t0d0 dosfs cfs this will create a dos partition and cfs partition total size of disk = 17836667 CISCO_UVFAT_1 vol /c0t0d0s1 part_off 63 part_siz 3906250 CISCO_BFS_1 vol /c0t0d0s2 part_off 3906323 part_siz 1024 CISCO_CFS_1 vol /c0t0d0s3 part_off 3907357 part_siz 13929300
It is also possible to repartition only the dosfs of disk 0 using the disk partition command, excluding the cfs keyword.
Console# disk partition /c0t0d0 dosfs total size of disk = 17836667 CISCO_UVFAT_1 vol /c0t0d0s1 part_off 63 part_siz 3906250 CISCO_BFS_1 vol /c0t0d0s2 part_off 3906323 part_siz 1024 CISCO_CFS_1 vol /c0t0d0s3 part_off 0 part_siz 0 Skipping creation of CFS partition for SCSI device
To repartition the disk 1, execute the disk partition command.
Console# disk partition /c0t1d0 cfs total size of disk = 17836667 CISCO_UVFAT_1 vol /c0t1d0s1 part_off 0 part_siz 0 Skipping creation of DOS partition for SCSI device CISCO_BFS_1 vol /c0t1d0s2 part_off 10 part_siz 1024 CISCO_CFS_1 vol /c0t1d0s3 part_off 1044 part_siz 17835613
Following a repartition, you need to format and mount the dosfs and cfs using the following commands:
If the first disk is reformatted, all the cache data is lost as well as all data in the dosfs partition. It will be necessary to reinstall the software image as well as the user interface images. See the section "Recovering the Cache Engine System Software."
If the second disk is reformatted, all the cache data on that disk (cfs) is lost. The Cache Engine will repopulate the disk with regular use.
Console# cli command Could not get stat information about /local <====
The dosfs partition may be corrupted if a power outage occurs while writing to the disk, or if the disk is removed while writing to the dosfs.
If there is a problem caused by the corrupted dosfs FAT table, perform the following steps:
Step 1 Execute the dosfs check /local command.
Step 2 If an error message occurs, execute the dosfs unmount /local command.
Step 3 Execute the dosfs repair auto /local command.
Step 4 Execute the dosfs mount /local command.
Step 5 If the dosfs mount fails after executing the dosfs repair command, execute the dosfs format command followed by the dosfs mount command.
If you are working in the current working directory of /local, you may see an error message (Could not get stat information about /local). You can execute the cd / command to avoid this error message.
Console# dosfs check /local tSyslog: Cannot write log to Disk File (/local/var/log/syslog.txt), OK
If a user password is forgotten, lost, or misconfigured, perform the following steps to reset the password:
Step 1 Attach an ASCII terminal to the Cache Engine console port.
Step 2 Break to the bootstrap program prompt and press any key before the Cache Engine completes the auto-boot sequence.
Enter c to change the auto-boot settings.
Enter net for boot device type.
Enter c to change the auto-boot settings.
Enter net for boot device type.
Enter c to change the auto-boot settings.
Enter flash for boot device type.
Press any key to stop auto-boot... 7 [PRESS ANY KEY NOW before the cache engine auto reboots] [CE500 Series]:c boot device (flash,net) [ net]: flash filename [ filename]: cache IP address [ 10.1.138.5]: IP network mask [ 255.255.255.0]: server IP address [ 171.69.63.242]: gateway IP address [ 10.1.138.1]: protocol (rcp,ftp,tftp) [ ]: username [ ]: password [ ]: flags [ 0]: 0x8000 [CE500 Series]:@
Step 3 Change the configuration flags field to 0x8000 (as shown above). Then enter @ to continue with the boot sequence.
Step 4 Once you get the CLI prompt, set the password for the user using the user modify username name password new-password command in the global configuration mode.
Step 5 Use the write command to save the configuration change.
By default, the Cache Engine has an `admin' user with no password set. When no password is set, you can enter anything as a password.
Use the following example configurations as a guide for setting up your Cache Engine. Remember that if a feature is not enabled, it will not appear in the running-configuration, though any settings configured for that feature will be included.
The following is an example WCCP Version 1 configuration:
Current configuration: ! ! logging event-export critical-events alert kernel logging 10.7.8.9 ! group add admin gid 0 group add everyone gid 1000 ! user add admin uid 0 password 1 "ceSzbyeb" capability admin-access user add britt uid 5001 password 1 "bbdze9eSbS" capability telnet-access ! ! ! hostname cache1D ! interface ethernet 0 ip address 10.1.1.34 255.255.255.0 ip broadcast-address 10.1.1.255 exit ! ! interface ethernet 1 exit ! ip default-gateway 10.1.1.2 ip name-server 10.2.2.2 ip domain-name cisco.com cron file /local/etc/crontab ntp server 10.6.14.7
! ! http proxy incoming 8080 icp server remote-client 10.4.6.8 no-fetch url-filter good-sites-allow wccp home-router 10.1.2.1 wccp version 1 ! transaction-logs destination disk transaction-logs interval 750 ! trusted-host 10.4.4.4 ! end cache1D#
The following is an example WCCP Version 2 configuration:
Building configuration... Current configuration: ! ! logging event-export critical-events alert kernel logging 10.7.8.9 ! group add admin gid 0 group add everyone gid 1000 ! user add admin uid 0 password 1 "ceSzbyeb" capability admin-access user add britt uid 5001 password 1 "bbdze9eSbS" capability telnet-access ! ! hostname cache1D ! interface ethernet 0 ip address 10.1.1.34 255.255.255.0 ip broadcast-address 10.1.1.255 exit
! ! interface ethernet 1 exit ! ip default-gateway 10.1.1.2 ip name-server 10.63.69.64 ip domain-name cisco.com cron file /local/etc/crontab ntp server 10.6.14.7 ! ! http proxy incoming 8080 http cluster max-delay 7 misses 200 icp client add-remote-server 10.4.1.8 parent icp-port 3130 http-port 8080 url-filter good-sites-allow wccp router-list 1 10.1.2.1 wccp router-list 3 10.3.1.1 10.4.2.1 wccp web-cache router-list 1 wccp version 2 wccp auth-bypass timer 3 ! transaction-logs destination disk transaction-logs interval 750 ! trusted-host 10.32.31.132 trusted-host 10.32.4.198 ! end cache1D#
You can remove a Cache Engine from a cache cluster or from a WCCP environment without losing any settings, and without powering off the unit. To remove a Cache Engine, disable the running version of WCCP by issuing the no wccp version 1 (or 2) command. The Cache Engine will still service proxy style requests.
Posted: Fri Jun 4 13:35:20 PDT 1999
Copyright 1989-1999©Cisco Systems Inc.