script/nextcloud
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fix: Vorzeitiger Abbruch bei großen Accounts nach 3600s Laufzeit

Browse Source 
Betroffen: scan_, recover_, restore_, recreate_bad_signature.sh
           und diagnose_share_key.sh

Bei Accounts mit sehr vielen bzw. sehr großen Dateien konnte der
jeweilige Pro-Account-Durchlauf länger als eine Stunde dauern und
wurde dann von PHP mit "Maximum execution time of 3600 seconds
exceeded" abgebrochen - mitten im Lauf, ohne jedes Ergebnis.

- su -c "$PHP_BIN ..." ruft PHP jetzt zusätzlich mit
  -d max_execution_time=0 auf.
- Das allein reicht nicht: Nextclouds eigenes lib/base.php setzt
  beim Bootstrap unbedingt (fest einprogrammiert, nicht
  konfigurierbar) set_time_limit(3600) und überschreibt damit den
  CLI-Flag wieder. Daher zusätzlich direkt nach dem require von
  lib/base.php ein erneutes set_time_limit(0) in jedem der fünf
  eingebetteten PHP-Scripte, das Nextclouds Reset seinerseits
  rückgängig macht.

Kein Verhaltensunterschied für kleine/normale Accounts, betrifft
nur die maximale Laufzeit pro Account.
This commit is contained in:
chris
2026-09-15 21:01:03 +02:00
parent 44f143fbe8
commit b33b859cf2
5 changed files with 69 additions and 23 deletions
Download Patch File Download Diff File Expand all files Collapse all files
recover_bad_signature.sh
+14 -8
View File
@@ -813,13 +813,8 @@ if ! $_revalidate_only_explicit && $terminal ; then
blank_line
echo -e "\033[37m\033[1mWhich mode should this run use?\033[m"
echo ""
echo -e " \033[1m[1] Recovery\033[m - decrypt bad-signature files (temporarily skips the signature check),
write them under \033[1m${DEFAULT_RECOVERY_BASE_DIR}/<website>\033[m and validate them"
echo ""
echo " [2] Revalidate-only - re-run just the validation checks against files a previous recovery run
already wrote to disk (same as '-V'); nothing is decrypted again and no config
value is touched"
echo -e " \033[1m[1] Recovery\033[m - decrypt bad-signature files (temporarily skips the signature check), write them under ${DEFAULT_RECOVERY_BASE_DIR}/<website> and validate them"
echo " [2] Revalidate-only - re-run just the validation checks against files a previous recovery run already wrote to disk (same as '-V'); nothing is decrypted again and no config value is touched"
info "Just press Return to use the default: [1] Recovery."
echo -n " Select mode by number [1]: "
read _mode_choice
@@ -1204,6 +1199,13 @@ if ($oldWorkingDir === false) {
}
chdir(__DIR__);
require_once __DIR__ . '/lib/base.php';
// Nextcloud's own bootstrap (lib/base.php) unconditionally calls
// set_time_limit(3600) as part of the require above - this OVERRIDES
// whatever -d max_execution_time was passed on the PHP command line,
// since it is a later, explicit runtime call. Undo that here, now that
// the require is done, so recovering a large account isn't killed
// after exactly one hour regardless of the CLI flag.
set_time_limit(0);
chdir($oldWorkingDir);
if (function_exists('posix_getuid') && posix_getuid() === 0) {
@@ -1438,7 +1440,11 @@ for _user in "${selected_user_arr[@]}" ; do
chown -R "$HTTP_USER":"$HTTP_GROUP" "$recovery_dir" 2> /dev/null
echononl " Recovering account \033[1;37m${_user}\033[m (${_user_total} files).."
su -c "$PHP_BIN $recovery_php_file $_user $list_file $user_out_dir" -s /bin/bash $HTTP_USER > "$user_result_tsv" 2> "$log_file"
# -d max_execution_time=0: decrypting/validating every file of a
# large account can run well over an hour; without this the PHP
# CLI process is killed by PHP's own execution-time limit
# (typically inherited from the webserver's php.ini) mid-run.
su -c "$PHP_BIN -d max_execution_time=0 $recovery_php_file $_user $list_file $user_out_dir" -s /bin/bash $HTTP_USER > "$user_result_tsv" 2> "$log_file"
_rc=$?
fi