This article goes over an issue seen when upgrading SRX1500 from 15.1X49 to 19.4R3 as per the upgrade path:

https://supportportal.juniper.net/s/article/Junos-upgrade-paths-for-SRX-platforms


This issue has also seen in post 19.4R3 releases when upgrading.

15.1X49 to 19.4R3 output:

root@SRX> request system software add /var/tmp/junos-srxentedge-x86-64-19.4R3-S11.2.tgz no-validate no-copy

Installing package '/var/tmp/junos-srxentedge-x86-64-19.4R3-S11.2.tgz' ...

Verified manifest signed by PackageProductionECP256_2022 method ECDSA256+SHA256

/dev/sdc1: open failed: No such device or address

/dev/sdc1: open failed: No such device or address

Saving contents of boot area prior to installation

sed: /var/tmp/preinstall/+COMMENT: No such file or directory

WARNING: This package will load JUNOS software.

WARNING: It will save JUNOS configuration files, and SSH keys

WARNING: (if configured), but erase all other files and information

WARNING: stored on this machine. It will attempt to preserve dumps

WARNING: and log files, but this can not be guaranteed. This is the

WARNING: pre-installation stage and all the software is loaded when

WARNING: you reboot the system.

Saving the config files ...

Pushing Junos image package to the host...

Installing /var/tmp/install-media-srx-700e-junos-19.4R3-S11.2.tgz

Extracting the package ...

tar: ./junos-srxentedge-x86-64-19.4R3-S11.2-linux.tgz: Wrote only 8192 of 10240 bytes

tar: Exiting with failure status due to previous errors

total 1149388

-rw-r--r-- 1 930 930 1051975853 Jan 23 18:31 junos-srxentedge-x86-64-19.4R3-S11.2-app.tgz

-rw-r--r-- 1 930 930 124985344 Jan 23 18:32 junos-srxentedge-x86-64-19.4R3-S11.2-linux.tgz

Setting up Junos host applications for installation ...

tar: ./srx-700e-19.4R3-S11.2.tgz: Wrote only 8192 of 9216 bytes

tar: Exiting with failure status due to previous errors

============================================

Host OS upgrade is FORCED

Current Host OS version: 1.0.0

New Host OS version: 3.0.4

Min host OS version required for applications: 0.2.4

============================================

Installing Host OS ...

cp: error writing '/var/tmp/./junos-srxentedge-x86-64-19.4R3-S11.2-linux.tgz': No space left on device

cp: failed to extend '/var/tmp/./junos-srxentedge-x86-64-19.4R3-S11.2-linux.tgz': No space left on device

upgrade_platform: -------------------

upgrade_platform: Parameters passed:

upgrade_platform: silent=0

upgrade_platform: package=/var/tmp/junos-srxentedge-x86-64-19.4R3-S11.2-linux.tgz

upgrade_platform: clean install=0

upgrade_platform: clean upgrade=0

upgrade_platform: Need reboot after staging=0

upgrade_platform: -------------------

upgrade_platform:

upgrade_platform: Checking input /var/tmp/junos-srxentedge-x86-64-19.4R3-S11.2-linux.tgz ...

gzip: stdin: unexpected end of file

tar: Unexpected EOF in archive

tar: Error is not recoverable: exiting now

upgrade_platform: Input package /var/tmp/junos-srxentedge-x86-64-19.4R3-S11.2-linux.tgz is not a valid .tgz file.

USAGE:

upgrade_platform [options] -p <package.tgz> , for upgrade

upgrade_platform -r , for rollback

[ options ]

-s | --silent : silent, no echo to terminal , just log

-n | --no-reboot : do not reboot after staging.

-r | --rollback : rollback the upgrade

-C | --clean-upgrade : like clean install

-I | --clean-install : like clean install, all partitions recreated

-c | --clear-state : clear all upgrade flags

-V | --validate : validate secure boot signatures

-h | --help : show this help

ERROR: Host Upgrade failed!

ERROR: upgrade failed

ERROR: junos-srx-img fails post-install

Installation failed for package '/var/tmp/junos-srxentedge-x86-64-19.4R3-S11.2.tgz'
 

On post 19.4R3 release you may find the following:

Installing Host OS ...
cp: error writing '/var/tmp/./junos-srxentedge-x86-64-23.2R1-S2.5-linux.tgz': No space left on device
cp: failed to extend '/var/tmp/./junos-srxentedge-x86-64-23.2R1-S2.5-linux.tgz': No space left on device
cp: error writing '/var/tmp/./upgrade_platform': No space left on device
cp: failed to extend '/var/tmp/./upgrade_platform': No space left on device

If you notice the device will not give you any errors, and will let you "upgrade" but once you reboot you will notice that the device is still on the same release.

Here are steps to take from less intrusive to most intrusive

Firstly, check current system storage with "show system storage" for anything out of the ordinary.

Here's an example of a relatively empty device on 15.1X49 to 19.4R3 scenario

root@SRX> show system storage 

--------------------------------------------------------------------------

Filesystem       Size    Used   Avail Capacity  Mounted on

/dev/vtbd0s1a      501M    342M    119M    74% /

devfs          1.0K    1.0K     0B   100% /dev

/dev/md0        980M    980M     0B   100% /junos

/cf           501M    342M    119M    74% /junos/cf

devfs          1.0K    1.0K     0B   100% /junos/dev/

procfs         4.0K    4.0K     0B   100% /proc

/dev/vtbd1s1e      1.6G    112K    1.4G    0% /config

/dev/vtbd1s1f      14G    3.5G    9.5G    27% /var

/dev/vtbd3s2       91M    914K    90M    1% /var/host

/dev/md1        320M    20M    275M    7% /mfs

/var/jail        14G    3.5G    9.5G    27% /jail/var

/var/jails/rest-api    14G    3.5G    9.5G   27% /web-api/var

/var/log         14G    3.5G    9.5G    27% /jail/var/log

devfs          1.0K    1.0K     0B   100% /jail/dev

192.168.1.1:/core    9.0G    21M    8.5G    0% /var/crash/corefiles

192.168.1.1:/var/volatile    7.8G    16K    7.8G  0% /var/log/host

192.168.1.1:/var/log    4.4G    2.1G    2.1G  50% /var/log/hostlogs

192.168.1.1:/var/traffic-log    50G    52M    47G  0% /var/traffic-log

192.168.1.1:/var/local    4.4G    2.1G    2.1G  50% /var/db/host

192.168.1.1:/var/db/aamwd    4.4G    2.1G    2.1G  50% /var/db/aamwd

192.168.1.1:/var/db/secinteld    4.4G    2.1G    2.1G  50% /var/db/secinteld

192.168.1.1:/app_disk    1.3G    2.0M    1.2G  0% /var/install_disk

Unexpected outputs would be "-5%" or "120%", these would suggest that there's something up with the internal storage.

If no unexpected outputs show there, you can do a cleanup of the device "request system storage", try again, and if that still does not work, check for large files in the device with:

>start shell 

% sh -c 'find / -size +10485760c 2> /dev/null' | xargs du -h | sort -nr

Here's an example output of 15.1X49 to 19.4R3:

root@> start shell 

% sh -c 'find / -size +10485760c 2> /dev/null' | xargs du -h | sort -nr

334M  /cf/packages/junos-srxjcp-15.1X49-D75.5-domestic

 29M  /modules/bcmsdk_5_9_x.ko

 22M  /usr/lib/dd/libjkernel-dd.tlv

 21M  /usr/sbin/rpd

 19M  /usr/sbin/ipfd

 17M  /modules/bcmsdk_5_6_1.ko

 14M  /var/rundb/schema.db

 14M  /usr/share/icu/4.6/icudt46l.dat

 14M  /usr/sbin/chassisd

 14M  /usr/sbin/authd

 14M  /usr/lib/dd/libjroute-dd.tlv

 13M  /usr/sbin/jpppd

 13M  /usr/sbin/aamwd

 12M  /usr/sbin/kmd

 12M  /usr/sbin/jdhcpd

 10M  /usr/lib/libidp-compiler.so.0

2.0G  /var/log/hostlogs/libdrv.log

1.2G  /var/tmp/preinstall/install-media-srx-700e-junos-19.4R3-S11.2.tgz

1.2G  /var/tmp/junos-srxentedge-x86-64-19.4R3-S11.2.tgz

If the largest file is the software package, then there's nothing much else to do, in this 15.1X49 example, pme could delete libdrv.log and the install-media file and try again.

On post 19.4R3 releases you may find a "previous_version_image.tgz", try deleting that one ("file delete" command).

If you deleted the files and there issue remains, or there were no large files, to begin with, on the 15.1X49 to 19.4R3 scenario you can open a JTAC case and ask for a new 19.4R3 version, just in case there's an issue with the file and you don't want to do the next step which is to reboot:

If you are just ok with it, or already asked JTAC for a new 19.4R3 image and the device keeps giving the same output, then reboot is the next step, this has proven to make the issue disappear.