Rex wrote: > Ok, that is odd now. Ah hah: https://www.ibm.com/docs/en/zos/2.5.0?topic=examples-example-3-packaging-sysmod-symbolic-link is an example showing: ++HFS(GSKAH010)
and says When SMP/E installs file GSKAH010 into the directory specified on the given DDDEF entry, the LINK and SYMLINK values will be resolved relative to the DDDEF directory. and sure enough, ls -i shows the same inode. So it's a hard link, and it's not actually taking space for the actual data, right? That I can live with. Might even note it in the doc, in case someone gets inquisitive and asks. Deleting the GSKAH010 version seems harmless (based on my testing-I can continue after doing so), but also pointless. I was mostly just concerned with leaving cruft that someone could ask about. Thanks to all-it was Shmuel's comments about LINK and SYMLINK that got me to this. I hadn't noticed SYMLINK in the doc, since it wasn't using it, and I was in fact not even thinking that LINK meant, well, a link-more just that it was some SMP/E jargon. But seeing SYMLINK Ied me to thinking "Well, if LINK means a hard link." Note that what I'm doing here is unwinding some undocumented work left by someone departed. It all works, but I want to understand it and optimize it as much as possible. I keep hoping that my fumbling through this stuff saves someone a few hours next week/month/year/decade. And, as ever, thanks for the assistance. You never know what offhand remark is going to lead to enlightenment! ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
