On 02/03/18 04:32 AM, George from the tribe wrote:
Is that something that is even possible?
There are a lot of possibilities. Do read this to start: https://www.ibm.com/developerworks/library/l-resizing-partitions-1/ <quote> Partition resizing is inherently at least a little bit risky. Partition resizing software must alter low-level partition and file system data structures and possibly move significant amounts of data. Problems such as corrupt data on the disk, power failures, and system crashes can all cause catastrophic failure of the resizing process. For this reason, you should not undertake a partition resizing operation lightly, and, whenever possible, you should back up your data, as described in "Preparing a backup." </quote> Consider also that you may need to run the partitioner "off line", that is from the recovery DVD or from a "LiveCD".
Does anyone have better recommendations? Like splitting off some of the directories from my root drive that take up a lot of memory, and moving them to the new partitions, and then just symlinking to those partitions in order to continue running?
Yes, you *can* but I wouldn't. It gets confusing come maintenance, and/or updates. if you do this, make sure that you document it. I'd use red ink. I really really really really REALLY hate pre-provisioning with a vengeance. I can't emphasise this too strongly. This approach to partitioning is a classic example of pre-provisioning. Ext4 is another example of pre-provisioning that I regularly rant against. That being said, paritioners have got smart over the years. See above. Once upon a time it took a tool called "Partition Magic" to shuffle partitions around in the way you are asking for, shrinking and growing file systems under the partitions in ways that the original FS designers never envisioned possible. Now we have file systems that can be grown if you can grow the partition. Check the man pages or Wikipedia. https://en.wikipedia.org/wiki/Comparison_of_file_systems#Resize_capabilities You can certainly grow an ext4FS, on-line or off-line. Again, check the man pages. You can't shrink one easily. back to the "smart" partitioner. As you can see, the Windows FAT file systems are not size-change friendly, but the NTFS is. Once upon a time after I'd moved my life to Linux I tried using the partitioner to shrink the Windows/FAT partition to make room for the Linux. I lost everything. I hope the partitioner tools have improved, but I'd still recommend making a complete backup. After that disaster I decided I really didn't need Windows on that machine. The one-and-only-one application I needed Windows for I could run on an older, lesser machine, when I needed to. I've never looked back. I've also never had to play repartitioning games because I also adopted LVM. This was more in my fight against pre-partitioning. I'd been using a similar tool under IBM's AIX on a an multi-processor cluster. That lets me shrink or grow a 'logical partition' on a live machine with the FS in use, or even extend the LP across drives or move to another drive. All the things you can do with BtrFS but this works with ANY file system. Of course if you do shrink of grow the LP you'll need a FS that shrinks and grows, too. That's why I chose to use ReiserFS. Not only is it a REAL B-Tree FS in that it has no pre-provisioning with a hard boundary between the number of i-nodes and the amount of data space, an archaic concept that goes back to the early 1970s and the old, slow, original UNIX v6/V7 file systems, but one that we can still see in ext4, but it can both shrink and grow on a life system. Other real B-tree file systems for Linux include, apart from BtrFS, XFS, JFS, and in due course, I hope, Reiser4FS. https://stackoverflow.com/questions/32512700/why-b-tree-for-file-systems The Linux JFS hasn't yet caught up with the version of JFS2 on AIX that I (occasionally) use in that it doesn't support both on-line grow and shrink. The Linux version I have to take off-line to grow. If you're happy with Ext4 and can put up with the pre-provisioning then OK. But even right now you might be facing a exhaustion of the data space in your root partition while there is plenty of inode space available. For example, on one of the few Ext4 file systems I have ... # df /usr/local Filesystem 1K-blocks Used Available Use% Mounted on /dev/mapper/vgmain-vLOCAL 3030800 1603412 1253720 57% /usr/local main:~ # df -i /usr/local Filesystem Inodes IUsed IFree IUse% Mounted on /dev/mapper/vgmain-vLOCAL 196608 11121 185487 6% /usr/local See: 57% of data space used but only 6% of inode space used. So I might get to the point were about 98% of the data space is used but only about 10% of the i-node space. Check your own system. That's why I prefer to use a REAL b-tree FS like JFS or XFS or ReiserFS. -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org