Opened 18 years ago

Closed 13 years ago

#148 closed defect (fixed)

parity issue with mdadm in mondorestore

Reported by: Bruno Cornec Owned by: Bruno Cornec
Priority: normal Milestone: 2.2.4
Component: mondo Version: 2.2.1
Severity: normal Keywords:
Cc: rbondoin@…

Description

mondorestore is unable to re-create a SW Raid5 with mdadm on RHEL 4 due to the call to mdadm which contains the -p option. Even if that option is documented, there seems to be no way to make it accept the parity option, and thus mondorestore fails. There is a need for a workaround in that case.

The one applied by the person reporting the bug to me was extreme, but kept for reference:

--- mondo-2.2.1/src/mondorestore/mondo-prep.c.ori       2006-08-05 12:00:51.000000000 +0200
+++ mondo-2.2.1/src/mondorestore/mondo-prep.c   2007-03-02 08:00:29.000000000 +0100
@@ -762,7 +762,8 @@
           "mdadm --create --force --run --auto=yes %s --level=%s --raid-devices=%d",
           raidlist->el[i].raid_device, level,
           raidlist->el[i].data_disks.entries);
-  if (raidlist->el[i].parity != -1) {
+/* RBN 02/03/2007 : bug parite / rhel4 => option mal interprete, suppression */
+/*  if (raidlist->el[i].parity != -1) {
     asprintf(&strtmp, "%s", program);
     paranoid_free(program);
     switch(raidlist->el[i].parity) {
@@ -783,7 +784,7 @@
       break;
     }
     paranoid_free(strtmp);
-  }
+  } */
   if (raidlist->el[i].chunk_size != -1) {
     asprintf(&strtmp, "%s", program);
     paranoid_free(program);

Change History (3)

comment:1 by Bruno Cornec, 18 years ago

Cc: rbondoin@… added
Status: newassigned

comment:2 by Bruno Cornec, 15 years ago

Milestone: 3.0.2

comment:3 by Bruno Cornec, 13 years ago

Milestone: 4.0.22.2.4
Resolution: fixed
Status: assignedclosed

This problem has been solved upstream in RHEL4.5 as per https://bugzilla.redhat.com/show_bug.cgi?id=199033 so closing here.

Note: See TracTickets for help on using tickets.