Submitted by: @pavel-zotov

Consider following script (from core-5392):

shell del c:\mix\firebird\qa\fbt-repo\tmp\c5392.fdb 2>nul;
create database 'localhost/3400:c:\mix\firebird\qa\fbt-repo\tmp\c5392.fdb';
shell gfix -w async localhost/3400:c:\mix\firebird\qa\fbt-repo\tmp\c5392.fdb;

commit;
connect 'localhost/3400:c:\mix\firebird\qa\fbt-repo\tmp\c5392.fdb';

create domain dm_longutf as varchar(8000) character set utf8;
recreate table test (id int not null, a int);
commit;

set echo on;

set term ^;
execute block as
declare i int;
declare n int = 100000;
begin
while (n>0) do insert into test(id, a) values(:n, :n) returning :n-1 into n;
end
^
set term ;^
commit;
alter table test add constraint pk_test primary key (id) using descending index pk_test_desc;
commit;

alter table test add b dm_longutf default '' not null;
commit;

update test set a=2;
rollback;

set list on;
-- this lead to decompression overran buffer (179), file: sqz.cpp line: 282
set echo on;
update test set a=3;
commit;

Statement "update test set a=3;" will lead to

Statement failed, SQLSTATE = XX000
internal Firebird consistency check (decompression overran buffer (179), file: sqz.cpp line: 282)

When this message appears one can NOT to stop FB service using control panel / services and FB process will load CPU on 100%.
Only ProcessExplorer kill command helps in this case.

Bug is reproducible on WI-T4.0.0.462 SuperServer.