On Thu, 23 Sep 2010, Robin Doherty wrote:
I have a RAID5 array of 5 1TB disks that has worked fine for 2 years but now says that it has 0 space available (even though it does have space available). It will allow me to read from it but not write. I can delete things, and the usage goes down but the space stays at 0. I can touch but not mkdir: rob@cholera ~ $ mkdir /share/test mkdir: cannot create directory `/share/test': No space left on device rob@cholera ~ $ touch /share/test rob@cholera ~ $ rm /share/test rob@cholera ~ $ Output from df -h (/dev/md2 is the problem array): Filesystem Size Used Avail Use% Mounted on /dev/md1 23G 15G 6.1G 72% / varrun 1008M 328K 1007M 1% /var/run varlock 1008M 0 1008M 0% /var/lock udev 1008M 140K 1008M 1% /dev devshm 1008M 0 1008M 0% /dev/shm /dev/md0 183M 43M 131M 25% /boot /dev/md2 3.6T 3.5T 0 100% /share and without the -h: Filesystem 1K-blocks Used Available Use% Mounted on /dev/md1 23261796 15696564 6392900 72% / varrun 1031412 328 1031084 1% /var/run varlock 1031412 0 1031412 0% /var/lock udev 1031412 140 1031272 1% /dev devshm 1031412 0 1031412 0% /dev/shm /dev/md0 186555 43532 133391 25% /boot /dev/md2 3843709832 3705379188 0 100% /share
Just a shot in the dark but I have seen this with Lustre systems. What does "df -i" show?
thanks -k -- To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
