# HG changeset patch
# User Henry S. Thompson <ht@inf.ed.ac.uk>
# Date 1704991396 0
# Node ID 6faed9e5d9c9d83329f8489dddf8e8428f8ba30a
# Parent  b6669d78a5d92d84a45ab1b559deb486c83f2ffe
use 2-digit suffixes,
use more care in what we wait for in the outer loop

diff -r b6669d78a5d9 -r 6faed9e5d9c9 bin/getcc_multi.aws
--- a/bin/getcc_multi.aws	Fri Dec 08 10:32:07 2023 +0000
+++ b/bin/getcc_multi.aws	Thu Jan 11 16:43:16 2024 +0000
@@ -40,9 +40,8 @@
 mkdir -p $s/orig/warc
 fgrep -v -f <(cd $s/orig/warc && ls *.warc.gz || :) <(fgrep -w $s warc.paths) > /tmp/hst/$s
 
-split -a 1 --numeric-suffixes=1 -n l/$nthreads /tmp/hst/$s /tmp/hst/${s}_
-
-seq 1 $nthreads | while read i
+split -a 2 --numeric-suffixes=1 -n l/$nthreads /tmp/hst/$s /tmp/hst/${s}_
+printf "%02d\n" $(seq 1 $nthreads) | while read i
 do
   cat /tmp/hst/${s}_$i | {
   printf "thread\t%s\t%s\t%s\n" $i $$ $(ps -o pgid= -p "$$") >> errlog_${SEG}_$i
@@ -57,5 +56,5 @@
   sleep 30
   }
 done
-while pgrep -a aws |grep -c s3; do sleep 60; done
+while pgrep -a aws |grep -c "aws s3.*${s}"; do sleep 60; done
 echo $(date) end $SEG