I am having tough time in getting time difference and size with uniform code like MB or GB or TB for entire client backup duration.

Below is my command:

mminfo -v -q "group=testgroup1,savetime>=02/17/2022,savetime<=02/18/2022" -r \
    "savetime,level,totalsize,volume,vmname,client,sscreate(20),sscomp(20)" -xc,

• -q is for query
• -r is to retrieve params from query
• group contains the clients
• savetime is day of backup
• level is level of backup
• totalsize is size of backup in bytes
• volume is the name of volume where data is stored
• vmname gives the name of vm
• client gives name of client , both vmname and client are one and same depends on type of backup.
• sscreate(20) gives start time of backup
• sscomp(20) gives end time of backup

o/p of my command looks like this

17/02/22,incr,4853101080,volume.001,,testclient1,17/02/22 20:27:18,17/02/22 20:40:45
17/02/22,incr,404305556,volume.001,,testclient1,17/02/22 20:27:15,17/02/22 20:27:34
17/02/22,incr,645786660,volume.001,,testclient1,17/02/22 20:27:17,17/02/22 20:27:30
17/02/22,incr,4,volume.001,,testclient1,17/02/22 20:27:45,17/02/22 20:27:47
17/02/22,incr,4,volume.001,,testclient1,17/02/22 20:27:16,17/02/22 20:27:19
17/02/22,incr,4,volume.001,,testclient1,17/02/22 20:27:46,17/02/22 20:27:48
17/02/22,incr,4,volume.001,,testclient1,17/02/22 20:28:05,17/02/22 20:28:08
17/02/22,incr,4,volume.002,,testclient1,17/02/22 20:27:48,17/02/22 20:27:51
17/02/22,incr,6085356,volume.002,,testclient1,17/02/22 20:42:26,17/02/22 20:42:51
17/02/22,incr,53328,volume.004,,testclient1,17/02/22 20:43:13,17/02/22 20:43:22
17/02/22,incr,4,volume.004,,testclient1,17/02/22 20:27:34,17/02/22 20:27:37

for each drive/mount point of a backup we have an entry in output in each line. Am able to sumup of size of data backed up per each client on a day but unable to get the logic for calculating time difference for backup for a client on one particular day. can someone help me with this?

logic which i used:(awk is where i started filtering the data)

printf "show name\n p type:nsr group\n" |
    nsradmin -i - |
    grep -v ^$ |
    cut -d: -f2- |
    cut -d\; -f1 |
    sort -u |
    perl -pe 's/\ //' |
    while read grp;do
        mminfo -q "group=$grp,savetime>02/12/2022 16:00,savetime<02/13/2022 16:00,level=full" \
             -r "savetime,level,totalsize,volume,vmname,client,sscreate(20),sscomp(20)" \
             -xc, 2>/dev/null |
           awk -F, 'BEGIN{OFS=FS}{if($5=="")$5=$6;else $5=$5;print}' |
           awk -F, '$5 != "vm_name" {a[$5","$2] =$3;b[$5","$2]  ;OFS=FS}
                     END{for (v in a) print v,b[v],a[v]}'|
           while read j;do echo $grp,$j;
       done
    done

o/p from code looks like this:

testgroup1,testclient1,full,11,65959975044

CodePudding user response：

Suggesting to fold all the post processing logic into a single gawk (standard Linux awk) script .

script.awk

function timestamp(dateStr) {
  formatedStr = gensub(/([[:digit:]]{2})\/([[:digit:]]{2})\/([[:digit:]]{2}) ([[:digit:]]{2}):([[:digit:]]{2}):([[:digit:]]{2})/, 
                       "20\\3 \\2 \\1 \\4 \\5 \\6", 1, dateStr);
  # from: DD/MM/DD HH:MM:SS
  # to:   YYYY MM DD HH MM SS
  return mktime(formatedStr);
}
BEGIN {
  minStartTime = 9999999999999999999999;
}
$5 == "" {
  $5 = $6;
}
{
  accumulatedDailyStorage[$5","$2]  = $3;
  accumulatedDailyTime[$5","$2]  = (timestamp($8) - timestamp($7));
  startTime = timestamp($7);
  minStartTime  = (startTime < minStartTime) ? startTime : minStartTime ; 
  endTime = timestamp($8);
  maxEndTime = (endTime > maxEndTime) ? endTime : maxEndTime ;
  accumulatedDailyCount[$5","$2]  ;
}
END {
  for (clientName in accumulatedDailyStorage) {
    print clientName, accumulatedDailyStorage[clientName], accumulatedDailyCount[clientName], accumulatedDailyTime[clientName], (maxEndTime - minStartTime);
  }
}

Output:

awk -F, -f script.awk input.1.txt
testclient1,incr 5909332004 11 889 6666