Upgrade issues after borked VM suspend

[samgreco]

I have been running ZCS FOSS on VMWare server for quite sometime now. And it has been running quite well, thank you. I have the host OS (Windows) do a complete backup copy once a week with a VMWare suspend. During the last suspend, something killed the process (OK, IT WAS ME! fat fingers...) then I couldn't start the server. Got that fixed and as is well. Mail is flowing in and out without issue. Except...

Seemed that now would be a good time to do the 5.0.18 update for the security patches, since I am still on 5.0.11. But when I try upgrade I get

[root@mail zcs]# ./install.sh --platform-override
./install.sh: line 26: ./util/utilfunc.sh: No such file or directory
./install.sh: line 29: ./util/modules/*sh: No such file or directory
./install.sh: line 90: ./util/globals.sh: No such file or directory
./install.sh: line 92: getPlatformVars: command not found
mkdir: missing operand
Try `mkdir --help' for more information.
chmod: missing operand after `750'
Try `chmod --help' for more information.

Operations logged to 
./install.sh: line 123: checkExistingInstall: command not found
./install.sh: line 134: displayLicense: command not found
./install.sh: line 136: checkUser: command not found
./install.sh: line 138: checkRequired: command not found
./install.sh: line 140: checkPackages: command not found
./install.sh: line 142: [: =: unary operator expected
./install.sh: line 184: loadConfig: command not found
./install.sh: line 186: checkVersionMatches: command not found
./install.sh: line 188: [: =: unary operator expected
./install.sh: line 203: removeExistingInstall: command not found
Installing packages

./install.sh: line 223: [: =: unary operator expected

Can anyone point me in the right direction? I have updated the system in the past without issue.

[uxbod]

Hmmm, what do you have in the directory where you have untar the install file ?

[samgreco]

[zimbra@mail zcs]$ ls
addUser.sh
admin.pdf
getconfig.sh
get_plat_tag.sh
globals.sh
Import_Wizard_Outlook.pdf
install.sh
MigrationWizard_Domino.pdf
MigrationWizard.pdf
OSmultiserverinstall.pdf
packages.sh
postinstall.sh
quick_start.pdf
readme_binary_en_US.txt
readme_source_en_US.txt
README.txt
RNZCSO_2005Beta.pdf
User Instructions for ZCS Import Wizard.pdf
utilfunc.sh
versions-init.sql
YPL.txt
zcl.txt
zcs-5.0.18_GA_3011.RHEL5.20090707164432.tgz
ZCS_Toaster.pdf
zimbra-apache-5.0.18_GA_3011.RHEL5-20090707164432.i386.rpm
zimbra-core-5.0.18_GA_3011.RHEL5-20090707164432.i386.rpm
Zimbra iCalendar Migration Guide.pdf
zimbra-ldap-5.0.18_GA_3011.RHEL5-20090707164432.i386.rpm
zimbra-logger-5.0.18_GA_3011.RHEL5-20090707164432.i386.rpm
zimbra-mta-5.0.18_GA_3011.RHEL5-20090707164432.i386.rpm
zimbra-proxy-5.0.18_GA_3011.RHEL5-20090707164432.i386.rpm
zimbra_public_eula_2.1.txt
Zimbra_Release_Note.pdf
Zimbra Schema.pdf
zimbra-snmp-5.0.18_GA_3011.RHEL5-20090707164432.i386.rpm
zimbra-spell-5.0.18_GA_3011.RHEL5-20090707164432.i386.rpm
zimbra-store-5.0.18_GA_3011.RHEL5-20090707164432.i386.rpm
zmdbintegrityreport
ZWC User Guide for 5_0.pdf

Should there be directories? Because all of the files are in the untarred directory. No sub directories.

Also, thanks for the amazingly fast response.

[uxbod]

Looks like it was unpacked incorrectly as it should look like

total 40
drwxrwxr-x 2 ssquid esets 4096 Jul  8 00:15 bin
drwxrwxr-x 2 ssquid esets 4096 Jul  8 00:15 data
drwxrwxr-x 3 ssquid esets 4096 Jul  8 00:15 docs
-rwxr-xr-x 1 ssquid esets 5919 Jul  8 00:15 install.sh
drwxrwxr-x 3 ssquid esets 4096 Jul  8 00:15 lib
drwxrwxr-x 2 ssquid esets 4096 Jul  8 00:15 packages
-r--r--r-- 1 ssquid esets  479 Jul  8 00:15 readme_binary_en_US.txt
-r--r--r-- 1 ssquid esets  428 Jul  8 00:15 README.txt
drwxrwxr-x 3 ssquid esets 4096 Jul  8 00:15 util

So move the .tgz out of the way and remove all those files. Once that is done untar the file again using tar zxvf .tgz.

[samgreco]

I guess the fat finger thing was worse than I thought.

No recursion can cause some strange stuff

Thanks uxbod

[uxbod]

I assume all is okay now and the thread can be closed ?

[samgreco]

I think so. I need to wait to do the upgrade until later when I can knock everyone out. BUT, I can tell it will be fine. I ran it and then killed it at the database integrity check.

Thanks again.