April 2024 - Libguestfs - libguestfs List Archives

what size should NBD_OPT_INFO report?
by Eric Blake 16 May '24

16 May '24

I noticed during integration testing that nbd-server blindly reports a size of 0 for all NBD_OPT_INFO requests, unless I pass a size argument on the command line to nbd-server. At first, I thought it was a side effect of me trying to use nbd-server on a block device (an LVM partition), as it is a common bug to rely on stat().st_size which only works for regular files (a block device has to use lseek(SEEK_END)); but then I noticed it happening when using nbd-server to serve regular files as well. I then turned to the source code, where I see that client->exportsize is set in just these places: commit_client() - exportsize = OFFT_MAX, then try setupexport() setupexport() - default to client->server->expected_size (if one was provided), further validating that actual size is large enough when actual size can be computed - if neither treefile or F_WAIT is set, compute actual size by opening one or more files and using size_autodetect() (which does the right thing for block devices, so my earlier thought about over-reliance on stat() was wrong) but these functions are only reached for NBD_OPT_EXPORT_NAME and NBD_OPT_GO, not NBD_OPT_INFO. The upshot is that for NBD_OPT_GO, there are some scenarios (treefile, F_WAIT) where nbd-server advertises a size of 9223372036854775807 (0x7fffffff_ffffffff) meaning unknown, but a size of 0 there is only possible if the file was successfully opened and really is zero bytes in length. Conversely, NBD_OPT_INFO is always advertising a size of 0, which means most of the time, the size changes between NBD_OPT_INFO and NBD_OPT_GO. For comparison, both nbdkit and qemu as an NBD server always advertise the same size for both NBD_OPT_INFO and NBD_OPT_GO (but it should also be noted that neither of these has the complexity of multifile like nbd-server). Should we tweak the NBD standard to recommend that a server advertise a size of OFFT_MAX for NBD_OPT_INFO if it is prohibitive to determine an actual size, rather than 0? Furthermore, is it worth adding code to make NBD_OPT_INFO try harder to provide a sensible value when possible (if expected_size was set, or if we are not multifile or F_WAIT, then a stat() is enough to get the size for serving a regular file; and if stat() says we have a block device, we can still try the open/lseek/close)? Also, is it worth trying to specify that since no known NBD servers allow exports with sizes larger than OFFT_MAX, a server MUST NOT report a size equal or larger than 9223372036854775808 (0x80000000_00000000)? There are definitely a few places that I could simplify in libnbd if we have a protocol guarantee that a valid export size will never have the most significant bit set, and therefore we don't have to worry about whether size is represented as a signed or unsigned value (which also implies that libnbd's function nbd_get_size() returning ssize_t instead of size_t is acceptable). But if we do that, then maybe 0x80000000_00000000 or even 0xffffffff_ffffffff would serve as a better recommended sentinel than 0x7fffffff_ffffffff for representing an indeterminate size. -- Eric Blake, Principal Software Engineer Red Hat, Inc. Virtualization: qemu.org | libguestfs.org

3 6

libnbd | Failed pipeline for master | 0ffa8366
by GitLab 29 Apr '24

29 Apr '24

Pipeline #1272215791 has failed! Project: libnbd ( https://gitlab.com/nbdkit/libnbd ) Branch: master ( https://gitlab.com/nbdkit/libnbd/-/commits/master ) Commit: 0ffa8366 ( https://gitlab.com/nbdkit/libnbd/-/commit/0ffa8366cfefda0ec60355f387d9b7425… ) Commit Message: Merge branch 'ci-proj' into 'master' ci: impor... Commit Author: Richard W_M_ Jones ( https://gitlab.com/rwmjones ) Pipeline #1272215791 ( https://gitlab.com/nbdkit/libnbd/-/pipelines/1272215791 ) triggered by Richard W_M_ Jones ( https://gitlab.com/rwmjones ) had 2 failed jobs. Job #6742108675 ( https://gitlab.com/nbdkit/libnbd/-/jobs/6742108675/raw ) Stage: builds Name: aarch64-macos-13 Job #6742108290 ( https://gitlab.com/nbdkit/libnbd/-/jobs/6742108290/raw ) Stage: builds Name: x86_64-alpine-edge -- You're receiving this email because of your account on gitlab.com.

1 0

nbdkit | Failed pipeline for master | 00d3d386
by GitLab 29 Apr '24

29 Apr '24

Pipeline #1272204598 has failed! Project: nbdkit ( https://gitlab.com/nbdkit/nbdkit ) Branch: master ( https://gitlab.com/nbdkit/nbdkit/-/commits/master ) Commit: 00d3d386 ( https://gitlab.com/nbdkit/nbdkit/-/commit/00d3d386c62e02b624484f29d75857484… ) Commit Message: docs: Fix references to nbdkit-sh-plugin(3) Th... Commit Author: Richard W_M_ Jones ( https://gitlab.com/rwmjones ) Pipeline #1272204598 ( https://gitlab.com/nbdkit/nbdkit/-/pipelines/1272204598 ) triggered by Richard W_M_ Jones ( https://gitlab.com/rwmjones ) had 2 failed jobs. Job #6742019448 ( https://gitlab.com/nbdkit/nbdkit/-/jobs/6742019448/raw ) Stage: builds Name: x86_64-freebsd-13 Job #6742019450 ( https://gitlab.com/nbdkit/nbdkit/-/jobs/6742019450/raw ) Stage: builds Name: x86_64-freebsd-14 -- You're receiving this email because of your account on gitlab.com.

1 0

[PATCH nbdkit] tests: gcs: skip tests without google cloud module
by Thomas Weißschuh 28 Apr '24

28 Apr '24

The gcs plugin uses the google cloud module and can't work without it. Skip the test if the module is unavailable. Signed-off-by: Thomas Weißschuh <thomas(a)t-8ch.de> --- tests/test-gcs.sh | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/test-gcs.sh b/tests/test-gcs.sh index ed823377dc5a..9b165e8159be 100755 --- a/tests/test-gcs.sh +++ b/tests/test-gcs.sh @@ -40,6 +40,7 @@ requires $PYTHON --version requires_nbdcopy requires_plugin python skip_if_valgrind "because Python code leaks memory" +requires $PYTHON -c 'import google.cloud' # Skip this test if the real google-cloud-storage module is installed, # since that module abuses pth files to insert itself into the module base-commit: ffa62d30fe6dcd29ae8e20df5c86d48505550614 -- 2.44.0

2 6

[PATCH libguestfs] test-data/binaries: Regenerate LoongArch files with stripping
by Richard W.M. Jones 25 Apr '24

25 Apr '24

These were regenerated by Dan Berrange from a known good Debian container image. In addition they are stripped. The commands were: $ podman run -it registry.gitlab.com/qemu-project/qemu/qemu/debian-loongarch-cross # echo 'int main(){}' > bin.c # loongarch64-unknown-linux-gnu-gcc bin.c -o bin-loongarch64 # loongarch64-unknown-linux-gnu-strip --strip-all bin-loongarch64 # echo '' > lib.c # loongarch64-unknown-linux-gnu-gcc -shared lib.c -o lib-loongarch64.so # loongarch64-unknown-linux-gnu-strip --strip-all lib-loongarch64.so Signed-off-by: Richard W.M. Jones <rjones(a)redhat.com> Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- test-data/binaries/bin-loongarch64-dynamic | Bin 20352 -> 18392 bytes test-data/binaries/lib-loongarch64.so | Bin 19320 -> 18064 bytes 2 files changed, 0 insertions(+), 0 deletions(-) diff --git a/test-data/binaries/bin-loongarch64-dynamic b/test-data/binaries/bin-loongarch64-dynamic index 645a5744a8eea77df243e5268264c2e0dcba0afb..5795de300d8b46a038723f0e00da047cb10816d6 100644 GIT binary patch delta 259 zcmXBPzbk}s00!{y^E$(I2IZC9ax#d)<uKfEdERfz-TTudWHsn!saxc&vnlcn$|9w# zr<cw0hQ(m9$f8?hG?^8j|KM|QfwyCvOpk3?^Jm!@<C(fC8qbu?_zi20yEl1xUHbVv zJ1jqR2G{$~JKLLgS=V`N;5+j<ilJt*HZxIMsLogCOs3{698*%>B1QzCn3DC0x>BE5 zRPGZ?LO`s@20RT4>IMyLDG!LY5E8qxA<<D968p+SqANtik!(bqD~*UN<q^>nV&Yyl pCZ3eW#EbHncoh=jT{a;;lqMD34h*Z8xGJo=#kTjK{ZhYG`32xgP-*}G delta 1856 zcmZ`(U1%It6#mXmHck3t{+n#8X{ry_VwR+lQlWD9Z+5eElazELT^~G|-JQhM{fRRZ zNQ$@-gD6TBuc8Zu)+faWX^9Ua3Vra+f|RNxAR_cdics_BTV!TuYPuP}m$~!Z?|kQ; zbI;7?`?PkIJ~%PBHVFrD01=!)Cn9(hM{kJnK>xjK7sdDWp36TUxqtSjuBSh~6Zw07 zV?FT0S=!u4ivb!;==qXX9T}epp9+tSKKqorlDFwtT>;@tplQAv3Idq-<rS>^@@@FS zmk;7kU%nk33hKLaI#93=-EDhlDCoetFW-r4zWjdN^yLrW)1c?y?JevC2XQ?pdzFOJ zg(VqPM$h$N<n{jq9u)NAgfHKRF<-tPkgqBO$^eG!r>rU`<(kq?L-MLJP;bNbj_2*6 z69DaM^@zLC2f$B13?P6*h_rhSP1tcZzuv-6xcqPJEmx3s5)b0yTGN1BqdAA)Tlj0P z(KGv8p^LX2{oV$CzNdgEn-Tw4*;kHHH#TKd>GSQ?caCCv=lX+NZQ@slKaBHkDIV^+ z`qsPp%Mo;8sihSAatH$$TEj`t!@&WvH@0_%S`r*5U_{=gzTk5N^xCV(ULc?whadFS zgXSsf1_3>I1Z6k9H@j~*k%O3T$?~opi9`6qEqmUze9P5)H|XI*SD$hHy^;FFJ}n^c zP*3m%0gubu)L*|)Fs$)#4a{1;#Fn;*f?=7~^1?#6fTC{bizU<24Yn#=DA%gG3Fcyr z7t6K0R_3BrGfdW&S5T-`mdd)N7sF%z3SKBxORO1&w#s_dGFGu*XqCFo#pOz670mXl z`^Kee$$Cg@>82GffH^rEol;qyPB4Sd&(6fv#N4dPne_CzXiDX5G?r1>K7m=|m$YiJ ztb-G|>F88C4(8PK9IKPA$z);{$C<NJanCQ6nTbU+Jd;dbP_wR=%(M;Hxm?N@*t9gm z;)+(PMpBt{EY7FG6Q{v?QL{8?`I=!p#HVN2w2GxFFPnN1tXGR*&YNb_2euD5ot<K@ z$=Rv6-Quj4w@=N570qs%RxS-+(sw%}Y_3);EswBi8J3pccsVjkbW}E?f}W6DQ9-9= zC?;r9=3;^pvJn$BEw^HVo|mDxpciE>F6b56hzq(Tx8j1<Whf!&9hplAx-1(BK_AJj zgrF-jqzd{{=2Ssn%Z4iGrrc5m-IAfCpkHJzDd>0END8_uw~~VH$xuqreVI!gqha}( z(kI_bbyL9pGDcS-%?XXTb9@BAD1Z)kaEIKd{YzBri2dKFP3lLI-$sXnJ>Ev=o2ll; LKTe8U(S!d2$&g3} diff --git a/test-data/binaries/lib-loongarch64.so b/test-data/binaries/lib-loongarch64.so index 92ba696580b3943bd6e4dad11d25b4acb18c8737..45dbdc1e2bb48612b174d20975543921336c0ccd 100644 GIT binary patch delta 1250 zcmZ`(T}V@L6hHraH$QUdbcHU(xjp1oa@o{ud)d94>`Tm6lw?8^B}|ZsNs<q#qt^<@ z0l|WZUJA*Ogh1$Hz6;??^bpir^fY@2lAy)^?i^V`=jC_L{h#wY=luR3KUc-dJ+b0d z?2-J*zNNGkOk)J2><a*>0stdi_Kn5A5tOoDL&$m3m3e8(OH&@>rX&Vn%zw_g|5u_w zp#{tdmoN8YIQn+@!;tUcBhR}#h2ZS&;x9?00zgff(=1$#0nutTr^0U{rq+Rh?tVwB z3N{#|VJl6ACd??!76%y1xUQVQ$4ZO!3;=jAAdcdm5?h*{8lTYSX0$#?45p&}nQpD8 zuaZK5AUWh$902UJ{3^d%C8bJ}&nn8MrS(!nNm7abf3lR<gUbMbXVpm|$8vPm3uxPT z(+KZ(ir$U$C1rP~sBBncaMxY%ei1O(AfT<7E8<sG6FJOU+OgMiTkLGTw1h)GF2>{} zui3CPXSpUAxE6MaByNW@VicYFx$%AKtC}Ll$Y<T3(f?Y)7Q)}9(v18xD858KSKcST zOT`EJe65|PPUCFEfrhYQ(5Sa1hA#O!0zF!{<_EkS3pppVH?{d|H|C~hvY=Hs8N44` zR{`oU6xW3Z7vj2T$L+W-ym%1Tg&#u+ml(zGQ3p;ZY)e4^vm-viUDPEbZz6TYC;kv) z0Oo|8&r>pXkP32-<;1KO*>SiCV3IQd%Q}YnL}%?)T8&~P*^D0(w$rj9CSi#uh+N(| zX{;fmA0wg}{mB~7^?xD_nv2mU@JhVhoux2tith}g1xMm8=PdxJr`sou54M4gsZ+-q zDpvStCvh*?>0Y7dfuufs7Wlc?#F3Ostl>gR7f*0IrHiL{kkZ8q3=QaF3l|1lqRHIB T=L2>OD2^j=fnKcFys>`(O-H%e literal 19320 zcmeI4PiP!f9LIk%lm1WJMAXKZ$`*Ur6gt*gF!rFECYv<4No|@^74c<ucap5^pRzmI zCLVN!3Oz)~#UAvcg`k3k;>n9TJt-*YQEv+N)?#x|6pJHo=J%bvPIs#nkM9G^Z{M5u zeg9|PyZnByXLDnwVGyuLO%Y}I!SC7V&q$azM4Kr?57Gl9m75@L2Oh$5g;KYfh+0oa zqSkA-Tc5l16}rYclXa`CTV);WSuPK|mHX616#iEEdD8CZa<H<z<m5H4DEs+u>EVLY z*Sw;vqtp0W+z(g`Q)=!fB8!N|r%uz$FYY+{4qa><Xny(q4?AZ|m%g|$DVvbbF}5xL zv<5t964?%Z_liZ^=%#EqfkoTt7A8!=q8=J*DH)7Ky_8AhKS@L>6j0dPAzzN6U3n4{ zGn1m|*4_D1!*lC1lOyF?)tzzjWjA`({)s3oIbyC<b;_kzUAb<&uQ1%7d3+ZV`74}{ zaiToZDITZXw|E64X(eCXUx$=@xvrpx^JN`sJ-ZAVxcpf!eQT_zFP{v|Z;W;G@}?}U z?DNw<8kE~@(7-QOZ>Db;lq1Sm1wSyGDSxobp#EmYH@gk$KR@f6D~@lT-%HQS^vYE~ z-Rh-GGv%iX8Q*Lfl*9iNQBG>1d(-jFwbfpd^?uwvL2I|NB<o+h(nGTTwXO+zZ-6pk zz3<<>-a;!dn;C!bT*@DGX8plt%QyEJ)PLPH4zvboG_2!|sE)AxpZs+2XmBQYEI8BM z)iSS}hSa5-89#ka_PynY{l%DNe=v^Vncx)WR-@5qG#ZUYqtR$I8jVJy(P%UpjYgx< zXfzs)Mx)VaG#ZUYqtR$I8jVJy(P%UpjYhKp{A%Gx`w%~Z|Lwz^hOcu!r{RY?ka!L^ z<WJe5`q#gLnk;gY)9{`BPCDc&e$hRg-X3Ms7~36e``E(2mlQu;37$*eZ3%pL_i*|{ z79zio(-_w(m)ph`_5Be<{g2!Ip9vUL>vKc!r|$&(o5x2+j;7M1ZocGHQ;$DkAGQx2 zd}@&YbrLfc8L=O}ibZC;ekvAO@j9wl)D^Fvibb2^byl%xb3Fel7Hx^wNyVbA@w}+q zyFdJncwehn)E)0<6^nMZT?dPH#rsdiqTSTT`%>jz{O~0f-KWY!<m=N#>+wDT-9%)z zi3_lc^N-+#WcG1B-ghAV{|zNB0Kcwbya}Vt{Jg^T;QIlZVLf-w^V=Qz-)UU$hV`_M zYtU{UJ_^gP2g&#5lQ93zb@@EZzcW6<*IW<YLvW4rlk59uhd2n^jrQ>o_IJpqLfvb4 zi*s{!fi|Ki2(KavGJZmXgeca;e7Tl)%A)Ai>J8y6E>WRYIa_u;w`d<~UqZ%IC<&)t zcb0`)_3FzsS9dC|C@xki%Or%75~XU%yRB%sl(!2c#Mt!kWL9LSMui|T`oh%k<irRG zF*7-WOi^sq#DY^T$|w)dO`jOaj-H;*irmE1v%}+AF*AHDmlY(&b0>}s=fsJzu~XTZ zIQRD04`Bd__CX@<5?X@n#&X4T@?<x>dSn;yuUhk5d%n78=NC)m;=xjp!qkG(SRlK& zTx~2@BJ0)3uDfMNX7azYWshvnUGm6w7oyoBdt;)=-k9I84yRHokUd}XBISI|BfC*_ zJcsOjqd|6|R;jpE58LWLBhO`dmhrI}f5#D51#whB;xj;pIgSPNIIm%kun;<&lStb{ z^ji?rpvU<Ri+cgeLl1_C!m}NEod2*s7eu@b=y5K~a5VC$gdX<^Slk}~zGvV#$D@HB z_W@Yk2a^3q`_rsH#uegz0gG}{FTcgf{+|iICs7O@3BEV5$+H`AZD3zY=n)qNwmUIk zj05CZkMSdZ3~aJ|GF|3z;a-IMBI4~}yZ~v~B_i6*Ybe2e@&d;dLOFmm?0KF$w2$vT z;u0bL5bm4MVgBCY@k5XMB;p)l{Q&f^7ZZA%3w<0XDH$6FI@suZi}tVt_qid~qZ~-4 zFY%s2{}In8(?Oru4)g){Ke^AKe*_IP(5n`EM`f1V$9o{}<MLaC{$pO{JsW*j**Y4f VGH8fNB&oN9{y4|Ux&mPb{oihP(+~gv -- 2.44.0

2 1

[PATCH libguestfs] test-data/binaries: Remove +x attribute
by Richard W.M. Jones 25 Apr '24

25 Apr '24

These binaries are not meant to be run, they are purely data files used for testing. Remove the +x attribute to prevent accidentally running them. However to avoid breaking the phony guests, we need to chmod +x the files when we upload them into those guests. --- test-data/binaries/bin-aarch64-dynamic | Bin test-data/binaries/bin-armv7-dynamic | Bin test-data/binaries/bin-i586-dynamic | Bin test-data/binaries/bin-ia64-dynamic | Bin test-data/binaries/bin-mipsel-dynamic | Bin test-data/binaries/bin-ppc64-dynamic | Bin test-data/binaries/bin-ppc64le-dynamic | Bin test-data/binaries/bin-riscv64-dynamic | Bin test-data/binaries/bin-s390x-dynamic | Bin test-data/binaries/bin-sparc-dynamic | Bin test-data/binaries/bin-win32.exe | Bin test-data/binaries/bin-win64.exe | Bin test-data/binaries/bin-x86_64-dynamic | Bin test-data/binaries/lib-aarch64.so | Bin test-data/binaries/lib-armv7.so | Bin test-data/binaries/lib-i586.so | Bin test-data/binaries/lib-ia64.so | Bin test-data/binaries/lib-mipsel.so | Bin test-data/binaries/lib-ppc64.so | Bin test-data/binaries/lib-ppc64le.so | Bin test-data/binaries/lib-riscv64.so | Bin test-data/binaries/lib-s390x.so | Bin test-data/binaries/lib-sparc.so | Bin test-data/binaries/lib-win32.dll | Bin test-data/binaries/lib-win64.dll | Bin test-data/binaries/lib-x86_64.so | Bin test-data/phony-guests/make-archlinux-img.sh | 1 + test-data/phony-guests/make-debian-img.sh | 1 + test-data/phony-guests/make-fedora-img.pl | 1 + test-data/phony-guests/make-ubuntu-img.sh | 1 + 30 files changed, 4 insertions(+) diff --git a/test-data/binaries/bin-aarch64-dynamic b/test-data/binaries/bin-aarch64-dynamic old mode 100755 new mode 100644 diff --git a/test-data/binaries/bin-armv7-dynamic b/test-data/binaries/bin-armv7-dynamic old mode 100755 new mode 100644 diff --git a/test-data/binaries/bin-i586-dynamic b/test-data/binaries/bin-i586-dynamic old mode 100755 new mode 100644 diff --git a/test-data/binaries/bin-ia64-dynamic b/test-data/binaries/bin-ia64-dynamic old mode 100755 new mode 100644 diff --git a/test-data/binaries/bin-mipsel-dynamic b/test-data/binaries/bin-mipsel-dynamic old mode 100755 new mode 100644 diff --git a/test-data/binaries/bin-ppc64-dynamic b/test-data/binaries/bin-ppc64-dynamic old mode 100755 new mode 100644 diff --git a/test-data/binaries/bin-ppc64le-dynamic b/test-data/binaries/bin-ppc64le-dynamic old mode 100755 new mode 100644 diff --git a/test-data/binaries/bin-riscv64-dynamic b/test-data/binaries/bin-riscv64-dynamic old mode 100755 new mode 100644 diff --git a/test-data/binaries/bin-s390x-dynamic b/test-data/binaries/bin-s390x-dynamic old mode 100755 new mode 100644 diff --git a/test-data/binaries/bin-sparc-dynamic b/test-data/binaries/bin-sparc-dynamic old mode 100755 new mode 100644 diff --git a/test-data/binaries/bin-win32.exe b/test-data/binaries/bin-win32.exe old mode 100755 new mode 100644 diff --git a/test-data/binaries/bin-win64.exe b/test-data/binaries/bin-win64.exe old mode 100755 new mode 100644 diff --git a/test-data/binaries/bin-x86_64-dynamic b/test-data/binaries/bin-x86_64-dynamic old mode 100755 new mode 100644 diff --git a/test-data/binaries/lib-aarch64.so b/test-data/binaries/lib-aarch64.so old mode 100755 new mode 100644 diff --git a/test-data/binaries/lib-armv7.so b/test-data/binaries/lib-armv7.so old mode 100755 new mode 100644 diff --git a/test-data/binaries/lib-i586.so b/test-data/binaries/lib-i586.so old mode 100755 new mode 100644 diff --git a/test-data/binaries/lib-ia64.so b/test-data/binaries/lib-ia64.so old mode 100755 new mode 100644 diff --git a/test-data/binaries/lib-mipsel.so b/test-data/binaries/lib-mipsel.so old mode 100755 new mode 100644 diff --git a/test-data/binaries/lib-ppc64.so b/test-data/binaries/lib-ppc64.so old mode 100755 new mode 100644 diff --git a/test-data/binaries/lib-ppc64le.so b/test-data/binaries/lib-ppc64le.so old mode 100755 new mode 100644 diff --git a/test-data/binaries/lib-riscv64.so b/test-data/binaries/lib-riscv64.so old mode 100755 new mode 100644 diff --git a/test-data/binaries/lib-s390x.so b/test-data/binaries/lib-s390x.so old mode 100755 new mode 100644 diff --git a/test-data/binaries/lib-sparc.so b/test-data/binaries/lib-sparc.so old mode 100755 new mode 100644 diff --git a/test-data/binaries/lib-win32.dll b/test-data/binaries/lib-win32.dll old mode 100755 new mode 100644 diff --git a/test-data/binaries/lib-win64.dll b/test-data/binaries/lib-win64.dll old mode 100755 new mode 100644 diff --git a/test-data/binaries/lib-x86_64.so b/test-data/binaries/lib-x86_64.so old mode 100755 new mode 100644 diff --git a/test-data/phony-guests/make-archlinux-img.sh b/test-data/phony-guests/make-archlinux-img.sh index 066df09d57..e0e3d3de7d 100755 --- a/test-data/phony-guests/make-archlinux-img.sh +++ b/test-data/phony-guests/make-archlinux-img.sh @@ -50,6 +50,7 @@ write /etc/hostname "archlinux.test" upload $SRCDIR/archlinux-package /var/lib/pacman/local/test-package-1:0.1-1/desc upload $SRCDIR/../binaries/bin-x86_64-dynamic /bin/ls +chmod 0755 /bin/ls mkdir /boot/grub touch /boot/grub/grub.conf diff --git a/test-data/phony-guests/make-debian-img.sh b/test-data/phony-guests/make-debian-img.sh index 637bd4bc88..27c68e6ece 100755 --- a/test-data/phony-guests/make-debian-img.sh +++ b/test-data/phony-guests/make-debian-img.sh @@ -85,6 +85,7 @@ write /etc/hostname "debian.invalid" upload $SRCDIR/debian-packages /var/lib/dpkg/status upload $SRCDIR/../binaries/bin-x86_64-dynamic /bin/ls +chmod 0755 /bin/ls upload $SRCDIR/debian-syslog /var/log/syslog diff --git a/test-data/phony-guests/make-fedora-img.pl b/test-data/phony-guests/make-fedora-img.pl index 6362e225b3..b7dc902e2b 100755 --- a/test-data/phony-guests/make-fedora-img.pl +++ b/test-data/phony-guests/make-fedora-img.pl @@ -331,6 +331,7 @@ $g->write ('/usr/lib/rpm/macros', <<EOF); EOF $g->upload ($ENV{SRCDIR}.'/../binaries/bin-x86_64-dynamic', '/bin/ls'); +$g->chmod (0755, '/bin/ls'); $g->tar_in ($ENV{SRCDIR}.'/fedora-journal.tar.xz', '/var/log/journal', compress => "xz"); diff --git a/test-data/phony-guests/make-ubuntu-img.sh b/test-data/phony-guests/make-ubuntu-img.sh index a3aa61050c..5e7a3a9686 100755 --- a/test-data/phony-guests/make-ubuntu-img.sh +++ b/test-data/phony-guests/make-ubuntu-img.sh @@ -76,6 +76,7 @@ write /etc/hostname "ubuntu.invalid" upload $SRCDIR/debian-packages /var/lib/dpkg/status upload $SRCDIR/../binaries/bin-x86_64-dynamic /bin/ls +chmod 0755 /bin/ls mkdir /boot/grub touch /boot/grub/grub.conf -- 2.44.0

1 0

[PATCH nbdkit v2 1/2] server: Reserve more characters from plugin/filter short names
by Richard W.M. Jones 22 Apr '24

22 Apr '24

The plugin/filter short name detection is very liberal, reserving only '.' and '/'. Thus, at least in theory, short plugin names containing almost arbitrary symbols and characters are permitted. Backslash ought to have been reserved when we added Windows support. We should probably reserve more characters, but in this commit I only reserve: * backslash (ie. directory separator on Windows) * ':' and ';' (common path separators) * '=' (used in nbdkit parameters) * space and comma (commonly used to separate lists) * non-printable ASCII characters Also DIR_SEPARATOR_STR, but that's likely to be already covered by the other tests so probably does nothing here. This commit is mainly about tightening up corner cases with possible security implications, for example if you managed to trick a program to invoke 'nbdkit "plugin param"' that might have an ambiguous parsing that you could use to your advantage. It should have no effect on normal, non-adversarial usage. --- server/options.h | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-) diff --git a/server/options.h b/server/options.h index 7d0730bae7..8b1bd679e9 100644 --- a/server/options.h +++ b/server/options.h @@ -117,7 +117,23 @@ static const struct option long_options[] = { static inline bool is_short_name (const char *filename) { - return strchr (filename, '.') == NULL && strchr (filename, '/') == NULL; + const size_t n = strlen (filename); + size_t i; + + for (i = 0; i < n; ++i) { + switch (filename[i]) { + case '\0'...31: case 127: /* non-printable ASCII */ + case '/': case '\\': /* directory separators */ + case ':': case ';': /* path separators */ + case ' ': + case '.': + case ',': + case '=': + return false; + } + } + + return strstr (filename, DIR_SEPARATOR_STR) == NULL; } #endif /* NBDKIT_OPTIONS_H */ -- 2.44.0

2 3

[PATCH nbdkit 1/2] server: Reject more characters in plugin/filter short names
by Richard W.M. Jones 22 Apr '24

22 Apr '24

The plugin/filter short name detection is very liberal, disallowing only '.' and '/'. Thus, at least in theory, short plugin names containing almost arbitrary symbols and characters are permitted. We should probably reserve more characters, but in this commit I only reserve: * backslash (ie. dir separator on Windows) * ':' and ';' (common path separators) * space * non-printable ASCII characters Also DIR_SEPARATOR_STR, but that's likely to be already covered by the other tests so probably does nothing here. This commit is mainly about tightening up corner cases with possible security implications, for example if you managed to trick a program to invoke 'nbdkit "plugin param"' that might have an ambiguous parsing that you could use to your advantage. It should have no effect on normal, non-adversarial usage. --- server/options.h | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) diff --git a/server/options.h b/server/options.h index 7d0730bae7..55e6abc28c 100644 --- a/server/options.h +++ b/server/options.h @@ -117,7 +117,21 @@ static const struct option long_options[] = { static inline bool is_short_name (const char *filename) { - return strchr (filename, '.') == NULL && strchr (filename, '/') == NULL; + const size_t n = strlen (filename); + size_t i; + + for (i = 0; i < n; ++i) { + switch (filename[i]) { + case '\0'...31: case 127: /* non-printable ASCII */ + case ' ': + case '.': + case '/': case '\\': /* directory separators */ + case ':': case ';': /* path separators */ + return false; + } + } + + return strstr (filename, DIR_SEPARATOR_STR) == NULL; } #endif /* NBDKIT_OPTIONS_H */ -- 2.44.0

2 4

[PATCH nbdkit 0/4] server: Improve errors from failed dlopen()
by Richard W.M. Jones 22 Apr '24

22 Apr '24

Improve errors from failed dlopen(), especially when the dlopen filename contains non-printable characters. Rich.

2 7

nbdkit | Failed pipeline for master | aade8812
by GitLab 20 Apr '24

20 Apr '24

Pipeline #1261201896 has failed! Project: nbdkit ( https://gitlab.com/nbdkit/nbdkit ) Branch: master ( https://gitlab.com/nbdkit/nbdkit/-/commits/master ) Commit: aade8812 ( https://gitlab.com/nbdkit/nbdkit/-/commit/aade881287fbf5c7ef3539d2f01e0d6ee… ) Commit Message: tests/test-shebang-crlf.sh: Fix location of she... Commit Author: Richard W_M_ Jones ( https://gitlab.com/rwmjones ) Pipeline #1261201896 ( https://gitlab.com/nbdkit/nbdkit/-/pipelines/1261201896 ) triggered by Richard W_M_ Jones ( https://gitlab.com/rwmjones ) had 4 failed jobs. Job #6675134840 ( https://gitlab.com/nbdkit/nbdkit/-/jobs/6675134840/raw ) Stage: builds Name: x86_64-freebsd-current Job #6675134839 ( https://gitlab.com/nbdkit/nbdkit/-/jobs/6675134839/raw ) Stage: builds Name: x86_64-freebsd-14 Job #6675134841 ( https://gitlab.com/nbdkit/nbdkit/-/jobs/6675134841/raw ) Stage: builds Name: aarch64-macos-14 Job #6675134838 ( https://gitlab.com/nbdkit/nbdkit/-/jobs/6675134838/raw ) Stage: builds Name: x86_64-freebsd-13 -- You're receiving this email because of your account on gitlab.com.

1 0