From 2a2353b20f2b5da84e6428c42b160e3a4d190b00 Mon Sep 17 00:00:00 2001
From: Carlos Salguero <carlos.salguero@percona.com>
Date: Thu, 25 Jul 2019 16:33:14 -0300
Subject: [PATCH] PT-1663 pt-stalk improvements

Implemented --retention-size and --retention-count
---
 bin/pt-stalk | 44 ++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 42 insertions(+), 2 deletions(-)

diff --git a/bin/pt-stalk b/bin/pt-stalk
index 2d65a88f..33adef2b 100755
--- a/bin/pt-stalk
+++ b/bin/pt-stalk
@@ -1352,6 +1352,8 @@ sleep_ok() {
 purge_samples() {
    local dir="$1"
    local retention_time="$2"
+   local retention_count="$3"
+   local retention_size="$4"
 
    # Delete collect files which more than --retention-time days old.
    find "$dir" -type f -mtime +$retention_time -exec rm -f '{}' \;
@@ -1363,6 +1365,30 @@ purge_samples() {
       find "$oprofile_dir" -depth -type d -name 'pt_collect_*' \
          -mtime +$retention_time -exec rm -rf '{}' \;
    fi
+
+   targetCnt=$(($retention_count + 0))
+   if [ $targetCnt -gt 0 ]; then
+      targetCnt=$(($retention_count + 1))
+      files_to_delete=$(find $dir -type f -exec basename {} \; | cut -f1 -d- | sort -r | uniq | tail -n +${targetCnt})
+      for prefix in $files_to_delete; do
+          echo "deleting files ${dir}${prefix}* according to the --retention-count param"
+          rm -f ${dir}${prefix}* 2>/dev/null
+      done
+   fi
+
+   targetSize=$(($retention_size + 0))
+   if [ $targetSize -gt 0 ]; then
+      files_to_delete=$(find $dir -type f -exec basename {} \; | cut -f1 -d- | sort -r | uniq | tail -n +1)
+      for prefix in $files_to_delete; do
+          current_size=$(du -BM $dir | cut -f1 -d"M")
+          if [ $current_size -gt $targetSize ]; then
+             echo "deleting files ${dir}${prefix}* according to the --retention-size param"
+             rm -f ${dir}${prefix}* 2>/dev/null
+          else
+             break
+          fi
+      done
+   fi
 }
 
 sigtrap() {
@@ -1471,7 +1497,7 @@ stalk() {
 
             # Purge old collect files.
             if [ -d "$OPT_DEST" ]; then
-               purge_samples "$OPT_DEST" "$OPT_RETENTION_TIME"
+               purge_samples "$OPT_DEST" "$OPT_RETENTION_TIME" "$OPT_RETENTION_COUNT" "$OPT_RETENTION_SIZE"
             fi
          fi
 
@@ -1496,7 +1522,7 @@ stalk() {
 
    # One final purge of old collect files, but only if in collect mode.
    if [ "$OPT_COLLECT" -a -d "$OPT_DEST" ]; then
-      purge_samples "$OPT_DEST" "$OPT_RETENTION_TIME"
+      purge_samples "$OPT_DEST" "$OPT_RETENTION_TIME" "$OPT_RETENTION_COUNT" "$OPT_RETENTION_SIZE"
    fi
 
    # Before exiting, the last collector may still be running.
@@ -2120,6 +2146,20 @@ The filename prefix for diagnostic samples.  By default, all files created
 by the same L<"--collect"> instance have a timestamp prefix based on the current
 local time, like C<2011_12_06_14_02_02>, which is December 6, 2011 at 14:02:02.
 
+=item --retention-count
+
+type: int; default: 0
+
+Keep the data for the last N runs. If N > 0, the program will keep the data for the last
+N runs and will delete the older data.
+
+=item --retention-size
+
+type: int; default: 0
+
+Keep up to --retention-size MB of data. It will keep at least 1 run even if the size is bigger
+than the specified in this parameter
+
 =item --retention-time
 
 type: int; default: 30