Generated Routes: English, please!

Unlike aggregate routes, the next hop associated with a generated route is the same next hop as that of the primary contributing route. The primary contributing route is the route with lowest route preference that falls within the aggregated range of prefixes. If there are multiple routes that fall within the aggregated range that share the same route preference, the route with the lowest number prefix, not the lowest prefix length, is selected as the primary contributing route.

The above is what I read about Generated routes. Seemed okay until I read below

1. Compare the protocol’s preference of the contributing routes. The lower the preference, the better the route. This is similar to the comparison that is done while determining the best route for the routing table.
2. Compare the protocol’s preference2 of the contributing routes. The lower preference2 value is better. If only one route has preference2, then this route is preferred.
3. The preference values are the same. Proceed with a numerical comparison of the prefixes' values.
   1. The primary contributor is the numerically smallest prefix value.
   2. If the two prefixes are numerically equal, the primary contributor is the route that has the smallest prefix length value.

But then it already made sense. Generated routes focus on a bigger network contrast that of an Aggregate route. Notice it mentioned "smallest prefix value" and "smallest prefix length" which are different.

Olive space allocation problem - WARNING: The /tmp/preinstall filesystem is low on free space.

I saw this in another blog:

WARNING: want 327680 for /tmp rather than 47328
Using 327680 for /tmp
Setting ospackage=jboot-9.3R4.4.tgz, configpackage=configs-9.3R4.4.tgz
Setting packlist=jbundle-9.3R4.4-export.tgz
WARNING: The /tmp/preinstall filesystem is low on free space.
WARNING: This package requires 202800k free, but there
WARNING: is only 162132k available.

WARNING: This installation attempt will be aborted. You are now
WARNING: in a debug shell. Type ^D to reboot the system
You are now in a debugging subshell (you may not see a prompt)...

Whoah!

No more space for the file system. I believe 2 years ago I just made the partitioning larger. I forgot if it solved my issue. But I do know I was at it for a month.

ha!

Most of you folks installing Olive first time would have had a problem installing it if you used any of the newer JunOS version other than 8.5, I think. (The -f was a quick fix)

My problem with a (relatively) newer JunOS version on QEMU is that it stops and complains of the same thing. Right now, I tried upping my VM's memory to 768Mb (company desktop, sssshhhhh!!!). Currently, it's doing this:

=================== Bootstrap installer starting ===================
Initialized the environment
Routing engine model is Olive
Discovered that flash disk = , hard disk = ad0
Disk to install is ad0
mfs: available=1442176
Using 1442176 for /tmp
Setting ospackage=jboot-11.1R6.4.tgz, configpackage=configs-11.1R6.4.tgz
Setting packlist=jbundle-11.1R6.4-domestic.tgz
Packages and configurations copied to /tmp
Filesystem 1K-blocks Used Avail Capacity iused ifree %iused Mounted on
/dev/md1 720876 360186 353482 50% 6 504 1% /tmp
Cleaning up ad0...
WARNING: Swap partition of ad0 is too small
WARNING: Need to partition ad0
Partitioning ad0 ...
******* Working on device /dev/ad0 *******
WARNING: Could not read disklabel for s1.
WARNING: Default sizes will be used for hard disk partitions.
Installing disk label on ad0s1
Running newfs on ad0s1a...
Reduced frags per cylinder group from 94048 to 94008 to enlarge last cyl group
/dev/ad0s1a: 921.0MB (1886204 sectors) block size 16384, fragment size 2048
using 6 cylinder groups of 183.61MB, 11751 blks, 23552 inodes.
super-block backups (for fsck -b #) at:
32, 376064, 752096, 1128128, 1504160, 1880192
Running newfs on ad0s1e...
/dev/ad0s1e: 102.0MB (208892 sectors) block size 16384, fragment size 2048
using 4 cylinder groups of 25.50MB, 1632 blks, 3328 inodes.
super-block backups (for fsck -b #) at:
32, 52256, 104480, 156704
Running newfs on ad0s1f...
/dev/ad0s1f: 79360.8MB (162530880 sectors) block size 16384, fragment size 2048
using 433 cylinder groups of 183.69MB, 11756 blks, 23552 inodes.

...and this as well...

super-block backups (for fsck -b #) at:
32, 376224, 752416, 1128608, 1504800, 1880992, 2257184, 2633376, 3009568,
3385760, 3761952, 4138144, 4514336, 4890528, 5266720, 5642912, 6019104,
6395296, 6771488, 7147680, 7523872, 7900064, 8276256, 8652448, 9028640,
9404832, 9781024, 10157216, 10533408, 10909600, 11285792, 11661984, 12038176,
... etc...

Till I got to this point...

Installing JUNOS on ad0...
Adding jbase...
Mounted jbase on /mnt/packages/mnt/jbase (/dev/md2)
Restoring backed up configurations...
Adding jbundle-11.1R6.4-domestic.tgz...
Checking package integrity...

.. and this too...

Running requirements check first for jbundle-11.1R6.4-domestic...
Running pre-install for jbundle-11.1R6.4-domestic...
Installing jbundle-11.1R6.4-domestic in /var/tmp/pa1803.44/jbundle-11.1R6.4-dome
stic.x1803...
Running post-install for jbundle-11.1R6.4-domestic...
Auto-deleting old jbase...

WARNING: A reboot is required to unload this software correctly
WARNING: Use the 'request system reboot' command
WARNING: when software installation is complete

WARNING: A reboot is required to load this software correctly
WARNING: Use the 'request system reboot' command
WARNING: when software installation is complete
Adding jbase...
Adding jkernel...
Adding jcrypto...
Adding jpfe...

Hmm, long wait till I get to know whether I should just knock my head to the nearest wall. :)