we can't do as leolaohoo mention,

Router required to remove first secondary ip address before primary ip address.

Can you not use loopback ip address either?

is the loopback address which known via static route with next hop of 10.205.30.26 will also be unreachable once the ip addres of the interface was removed?

Gee ... Ok.

Presume you got this ...

int f 0/0

ip address 1.1.1.1 255.255.255.0

ip address 1.1.2.1 255.255.255.0 secondary

and you want to delete the first secondary, so you put another network there like "ip address 1.1.3.1 255.255.255.0", get out and telnet into the 1.1.3.1 and do a "no ip address 1.1.2.1 255.255.255.0 secondary".

Otherwise, create a loopback address and telnet into that and put the 1.1.3.1 secondary as well as remove the 1.1.2.1 address. Does this make sense to you?

but is the loopback address which is known only through the ip address of 10.205.30.26 will also be unreachable once the ip address of the interface was removed?

I am still getting likes for the above, but this is actually a quite dated approach. A couple other options exist that are more viable:

1. Put all configuration changes you need to make into a text file on flash memory, just as if you were going to copy and paste them into the device.  You can use TCL to write directly to flash memory as follows:

tclsh

puts [open flash:changes.txt w+] {
<insert changes here just as if you were typing into config mode>
}

tclquit

Then, when ready to execute, just copy from flash to running config as follows:

copy flash:changes.txt running-config

This will perform a merge of the changes in the text file with the existing running configuration.  As always, I recommend scheduling a reload and removing VTY line ACLs and AAA as a safety precaution.

2. Similar to the above, the configuration changes can be put into an EEM script that is triggered with "event none", which means it must be manually run by the engineer.  Then you just run the script.  This is more useful in cases where you may need to make and revert changes regularly.  For example, in an MPLS L3VPN environment, I have an EEM script that modifies BGP policy to route a given location to its alternate internet access point, and a second EEM script to revert those changes.  I keep them on the device so that I can regularly execute re-routes for various testing scenarios (such as transparent cyber security systems at alternate egress locations, etc). Just do a little googling to get the syntax of the EEM script, but the most critical element is to disable command authorization for your script and to ensure you understand the cli for EEM needs to elevate (enable) and enter configuration mode (conf t) before you execute the commands you want to do (i.e. the first two lines of the EEM should include 'cli command "enable"' and 'cli command "config t"').  You'll get the hang of it quickly, just do some testing of your EEM before deploying (run CSR1000v in VirtualBox or something like that).